system

Abstract

We provide the system. [Solution] A means for receiving user input, A means of analyzing input data, A method using a generative artificial intelligence that generates a response based on the analysis results, A means of sending the generated response to the NPC, A means by which NPCs respond to users, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the current metaverse environment, the responses of NPCs (non-player characters) that interact with users are based on preset fixed scripts, making it difficult to flexibly respond to users' intentions and questions. For this reason, there is a problem that the user experience is restricted within a certain range, and information provision and purchase promotion required by users cannot be sufficiently carried out.

Means for Solving the Problems

[0005] To solve the aforementioned problems, the present invention provides the following means: a system including means for receiving user input, means for analyzing user input data, means for using a generative artificial intelligence to generate a response based on the analysis results, means for sending the generated response to an NPC, and means for the NPC to respond to the user. The system may also include means for speech recognition to convert user voice input into text, and means for using a natural language processing engine as an analysis means. This enables flexible response generation to user input and improves the user experience within the metaverse.

[0006] "User input" refers to the voice or text data that a user provides when interacting with an NPC within the metaverse.

[0007] "Input data analysis" is the process of understanding voice or text data obtained from users and analyzing the user's intent and the content of their questions.

[0008] "Generative artificial intelligence" is a type of artificial intelligence technology that generates natural language responses based on user input data.

[0009] An "NPC (Non-Player Character)" is a virtual character that interacts with the user within the metaverse, and its behavior and dialogue are controlled by the system.

[0010] "Voice recognition means" refers to technologies and devices that analyze voice input from a user and convert it into text data.

[0011] A "natural language processing engine" is software that analyzes text data, understands human language, and extracts appropriate responses and information. [Brief explanation of the drawing]

[0012] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2]This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, when an emotion engine is combined. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, a tagged processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, a tagged RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, a tagged storage is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0018] In the following embodiments, a tagged communication I / F (Interface) is an interface that includes a communication processor, an antenna, and the like. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] This invention relates to a system that improves the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment.

[0034] Overall system configuration

[0035] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. The following describes each part of the system, subject by subject.

[0036] User-side operations

[0037] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0038] Operation on the device side

[0039] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0040] Server-side operations

[0041] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0042] The generative AI generates appropriate text responses in response to user input and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0043] NPC-side operations

[0044] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0045] Specific example

[0046] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0047] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0048] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0049] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I'm looking for shoes." Based on the analysis results, it instructs a generative AI to generate a response such as, "We have new shoes here. Would you like to take a look?"

[0050] 4. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0051] 5. NPC: Says to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[0052] 6. User: Follow the NPC's instructions and go to check out the new shoes.

[0053] In this way, the present invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, the user can achieve their goals more intuitively and effectively.

[0054] The following describes the processing flow.

[0055] Step 1:

[0056] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0057] Step 2:

[0058] The device captures the user's voice or text input. If voice input is received, the voice data is collected and prepared for processing.

[0059] Step 3:

[0060] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0061] Step 4:

[0062] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0063] Step 5:

[0064] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0065] Step 6:

[0066] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0067] Step 7:

[0068] The server sends a response generation request to the generative AI based on the analysis results of the NLP engine. The request is created based on the contextual information derived from the analysis results.

[0069] Step 8:

[0070] The generative AI receives a request from the server and generates an appropriate response. The generated response is returned to the server in text format.

[0071] Step 9:

[0072] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0073] Step 10:

[0074] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0075] Step 11:

[0076] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0077] Step 12:

[0078] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0079] Through the processing flow described above, the system of the present invention enables natural interaction and guidance with the user within the metaverse.

[0080] (Example 1)

[0081] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0082] In traditional metaverse environments, interactions with NPCs (non-player characters) were based on pre-programmed scripts, which presented challenges in adequately responding to the diverse questions and requests users might make. Furthermore, natural, real-time dialogue was difficult, limiting the user experience. As a result, users lacked flexible guidance for their actions within the metaverse, leading to reduced usability.

[0083] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0084] In this invention, the server includes means for acquiring user voice or text input, speech recognition means for converting voice input into text data, means for using a natural language processing engine to analyze the acquired text data, means for using generative artificial intelligence to generate an appropriate response based on the analysis results, means for converting the generated response into an NPC action script and sending it to the NPC, and means for the NPC to respond to the user. This makes it possible to respond flexibly to a variety of user questions and requests in real time, realize more natural and intuitive dialogue within the metaverse, and improve the user experience.

[0085] "Means for acquiring user voice or text input" refers to means for recognizing voice or text input made by a user within the metaverse and capturing it for processing within the system.

[0086] A "speech recognition means for converting voice input into text data" is a means that uses speech recognition technology to analyze a user's voice input and convert it into corresponding text data.

[0087] "Methods using a natural language processing engine to analyze acquired text data" refers to methods that utilize natural language processing technology to analyze generated text data and understand the user's intent and the content of the question.

[0088] "Means of using generative artificial intelligence to generate appropriate responses based on analysis results" refers to means of using generative artificial intelligence technology to generate appropriate responses for the user based on analysis results obtained from a natural language processing engine.

[0089] "Means for converting generated responses into action scripts for NPCs and sending them to NPCs" refers to a means of converting responses generated by a generative artificial intelligence into action scripts that can be executed by NPCs and sending them to NPCs.

[0090] "Means by which NPCs respond to users" refers to means by which NPCs provide responses to users by performing voice output or gestures based on behavioral scripts.

[0091] This invention relates to a system that enhances the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment. The following describes in detail the specific forms for implementing this system.

[0092] Overall system configuration

[0093] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and provides them to the user through NPCs. The system's components are as follows:

[0094] 1. User-side operations

[0095] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0096] 2. Operation on the device side

[0097] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0098] 3. Server-side operations

[0099] The server uses a natural language processing (NLP) engine to analyze the text data received from the terminal. The NLP engine (e.g., spaCy or NLTK) understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model, such as OpenAI's GPT-3®) to generate an appropriate response. The generative AI generates an appropriate response in text form for the user's input and returns it to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0100] 4. Actions taken by the NPC

[0101] NPCs respond to users according to behavioral scripts received from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed (e.g., pointing to a product shelf).

[0102] Specific example

[0103] Below is an example of a specific user-NPC interaction in a fashion shop within the metaverse.

[0104] 1. The user enters a fashion shop in the metaverse and speaks to an NPC saying, "I'm looking for new shoes."

[0105] 2. The device captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0106] 3. The server analyzes the received text data using an NLP engine to understand the user's intent, "I'm looking for shoes." Based on this analysis, it sends a prompt message to the AI ​​model: "The user is looking for new shoes. Please guide them to our products as an appropriate response." This prompt generates the appropriate response: "We have new shoes here. Would you like to take a look?"

[0107] 4. The server converts the generated response into an NPC action script and sends it to the NPC. The action script includes actions such as "talk" and "point to the product shelf".

[0108] 5. The NPC responds to the user with a voice message saying, "We have new shoes here. Would you like to take a look?" and then points to them.

[0109] 6. The user follows the NPC's instructions and goes to check out the new shoes.

[0110] In this way, this invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, users can achieve their goals more intuitively and effectively.

[0111] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0112] Step 1:

[0113] The user speaks to an NPC in the metaverse via voice or text. For example, the user might say, "I'm looking for new shoes." The input is the user's voice or text, and the output is captured on the user's device.

[0114] Step 2:

[0115] The device converts the captured audio data into text data using speech recognition. This process uses a speech recognition API (e.g., a speech recognition service) to convert the audio "I'm looking for new shoes" into text data with the same meaning. The text data is generated as output and passed to the next step.

[0116] Step 3:

[0117] The terminal sends the converted text data to the server. This communication uses a secure method such as the HTTPS protocol. The input is the text data generated in step 2, and the output is the text data transferred to the server.

[0118] Step 4:

[0119] The server uses a natural language processing (NLP) engine (e.g., spaCy or NLTK) to parse the received text data. Specifically, it parses the text "I'm looking for new shoes" and understands that the user's intent is "I'm looking for shoes." This parsing process is a data processing step that breaks down text data into meaning and generates the result. The input is text data, and the output is the parsed result regarding the user's intent.

[0120] Step 5:

[0121] The server sends a prompt to a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. As an example of a prompt, it generates the sentence "The user is looking for new shoes. Please guide them to products as an appropriate response" and sends it to the AI ​​model. The input is a prompt based on the analysis results, and the output is a request to the generative AI model.

[0122] Step 6:

[0123] The generative AI model generates an appropriate response based on the given prompt. For example, it generates the text response, "We have new shoes here. Would you like to take a look?" The input is the prompt, and the output is the generated response text.

[0124] Step 7:

[0125] The server converts the generated response text into an NPC action script. This action script includes actions such as speaking the response and specific gestures (e.g., pointing to a product shelf). The input is the generated response text, and the output is the action script.

[0126] Step 8:

[0127] The server sends an action script to the NPC. This prepares the NPC to provide a response to the user. The input is the action script, and the output is the script sent to the NPC.

[0128] Step 9:

[0129] The NPC responds to the user according to the action script received from the server. Using a voice output device, it might say, "We have new shoes here. Would you like to take a look?" and simultaneously point to them. The input is the action script, and the output is the response and action to the user.

[0130] (Application Example 1)

[0131] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0132] In metaverse environments, user-NPC dialogue systems are limited to standard text-based responses and simple preset responses, resulting in a restricted user experience. Virtual stores, in particular, require a concrete dialogue system that allows users to easily obtain product information and enjoy intuitive shopping. Conventional systems lack real-time natural language response, speech recognition, and speech output capabilities, often leading to a lower quality user experience.

[0133] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0134] In this invention, the server includes means for receiving user input, means for analyzing input data, means for using generative artificial intelligence to generate a response based on the analysis results, means for transmitting the generated response to a display character, means for the display character to respond to the user, and means for voice recognition and voice synthesis for voice input and voice output. This enables users visiting a virtual store to engage in natural and intuitive real-time conversations with an NPC using a head-mounted display, and to smoothly acquire product information and proceed with purchase procedures.

[0135] "User input" refers to actions and information provided by users in the metaverse environment via voice or text.

[0136] "Input data" refers to audio and text data obtained from user input.

[0137] "Analysis" is the process of understanding input data and interpreting the user's intent and the content of their questions.

[0138] "Generative artificial intelligence" is an AI technology that generates appropriate responses based on the results of analyzing input data.

[0139] A "display character" refers to an NPC (non-player character) that responds to the user in the metaverse environment.

[0140] "Speech recognition" is a technology that converts a user's voice input into text data.

[0141] "Speech synthesis" is a technology that converts text data into speech data and outputs it as speech.

[0142] "Natural language processing" refers to a set of computational techniques and methods for understanding, analyzing, and generating human language.

[0143] This invention provides a system that allows users to interact with NPCs in real time using a head-mounted display within a virtual store. Users can efficiently obtain product information and proceed smoothly with the purchase process. The configuration and processing of each part of the system are shown below.

[0144] User actions

[0145] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs) using their voice. For example, a user might say, "I'm looking for a new smartphone." This input is captured by a microphone built into the user's head-mounted display. The voice data is then converted into text data within the device.

[0146] Terminal operation

[0147] The terminal uses speech recognition software to convert the user's voice input into text data. The speech recognition software used is the Python `speech_recognition` library. The converted text data is sent to the server.

[0148] Server Operations

[0149] The server analyzes the received text data using a natural language processing engine (NLP engine). The NLP engine used is the nlptown / bert-base-multilingual-uncased-sentiment model. Based on the analyzed user intent and question content, it queries a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. This generated response text is then sent back to the user's terminal by the server and converted into an action script for the displayed character.

[0150] NPC Controls

[0151] The NPC displayed on the head-mounted display uses text-to-speech software to play back response text received from the server. The pyttsx3 library is used for this purpose. The NPC provides the generated response to the user as audio, and, if necessary, accompanies it with body movements.

[0152] Specific example

[0153] When a user says "I'm looking for a new smartphone" in the virtual store, the system responds as follows:

[0154] Voice input: "I'm looking for a new smartphone."

[0155] Analysis of user intent: "Prefers to view on a smartphone"

[0156] Response generation: "We have a new smartphone here. Would you like to take a look at the latest model?"

[0157] Example of a prompt

[0158] After analyzing the user's voice input as "I'm looking for a new smartphone," the following prompt is entered into the generative AI model:

[0159] "Generate appropriate responses for users looking for a new smartphone."

[0160] This provides a convenient system that allows users to intuitively obtain product information and effectively proceed with the purchase process.

[0161] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0162] Step 1:

[0163] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs). During this process, the user's voice input is captured by a microphone built into the head-mounted display.

[0164] Input: User voice input (e.g., "I'm looking for a new smartphone.")

[0165] Output: Captured audio data

[0166] Step 2:

[0167] The device converts the captured audio data into text data using speech recognition software. This speech recognition uses the Python `speech_recognition` library.

[0168] Input: Captured audio data

[0169] Data processing: Text conversion using speech recognition.

[0170] Output: Converted text data (e.g., "I'm looking for a new smartphone.")

[0171] Step 3:

[0172] The terminal sends the converted text data to the server.

[0173] Input: Converted text data

[0174] Output: Text data sent to the server

[0175] Step 4:

[0176] The server analyzes the received text data using a natural language processing engine. This engine uses the nlptown / bert-base-multilingual-uncased-sentiment model.

[0177] Input: Received text data

[0178] Data processing: Analyzing user intent using natural language processing

[0179] Output: Analyzed user intent (e.g., "I want to view this on my smartphone")

[0180] Step 5:

[0181] The server sends a prompt to a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. The prompt is "Generate an appropriate response for a user looking for a new smartphone."

[0182] Input: Analyzed user intent, prompt message

[0183] Data processing: Prompt-based response generation

[0184] Output: Generated response text (Example: "We have a new smartphone here. Would you like to take a look at the latest model?")

[0185] Step 6:

[0186] The server sends the generated response text to the user's terminal and converts it into a character script for display on the terminal.

[0187] Input: Generated response text

[0188] Output: Converted display character script

[0189] Step 7:

[0190] The terminal uses the converted script to have the displayed character (NPC) play a response using speech synthesis software (e.g., pyttsx3). During this process, speech synthesis generates audio data, which is then provided to the user through the NPC's speaker. The NPC also performs gestures and other actions as needed.

[0191] Input: Converted display character script

[0192] Data processing: Generation of voice data using speech synthesis, and operation instructions based on scripts.

[0193] Output: Voice responses and gestures to the user

[0194] Through these steps, the user experience within the virtual store is significantly improved, allowing users to intuitively obtain the necessary information and proceed with the purchase process smoothly.

[0195] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0196] This invention relates to a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining a generative artificial intelligence system and an emotion engine for NPCs (non-player characters).

[0197] Overall system configuration

[0198] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. Furthermore, by incorporating an emotion engine, it recognizes the user's emotional state and generates responses accordingly. The following describes each part of the system, subject by subject.

[0199] User-side operations

[0200] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0201] Operation on the device side

[0202] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0203] Server-side operations

[0204] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0205] Furthermore, the server uses an emotion engine to analyze the user's voice or text data and recognize the user's emotions. Based on the analysis results of the emotion engine, the generative AI adjusts its response. This process generates a response that is more appropriate to the user's state of mind.

[0206] The generative AI generates appropriate text responses corresponding to user input and the results of the emotion engine, and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0207] NPC-side operations

[0208] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0209] Specific example

[0210] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0211] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0212] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0213] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I am looking for shoes." Furthermore, it uses an emotion engine to analyze the user's voice data and recognize emotional states such as "excited" or "distressed."

[0214] 4. Server: Based on the analysis results and the emotion engine's results, it gives instructions to the generative AI to generate responses such as, "We have new shoes here. Would you like to take a look?" or "I'm here to help if you need anything."

[0215] 5. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0216] 6. NPC: The NPC will say to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf. It can also change its facial expressions and tone of voice in response to the user's emotions.

[0217] 7. User: Follow the NPC's instructions and go to check out the new shoes.

[0218] In this way, the system of the present invention improves the user experience within the metaverse. By combining it with an emotion engine, NPCs can not only respond flexibly to the user's specific questions and intentions, but also provide responses that take into account the user's emotional state.

[0219] The following describes the processing flow.

[0220] Step 1:

[0221] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0222] Step 2:

[0223] The device captures the user's voice or text input. In the case of voice input, it collects the voice data and prepares it for processing.

[0224] Step 3:

[0225] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0226] Step 4:

[0227] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0228] Step 5:

[0229] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0230] Step 6:

[0231] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0232] Step 7:

[0233] The server invokes an emotion engine to analyze the received audio or text data and recognize the user's emotions. For example, it extracts emotions from voice tone or text.

[0234] Step 8:

[0235] The server sends a request to the generative AI to generate a response based on the analysis results of the NLP engine and the emotion engine. For example, it provides context such as "the user is looking for shoes" or "the user is in trouble."

[0236] Step 9:

[0237] A generative AI receives a request from a server and generates an appropriate response. For example, it might generate a response like, "We have new shoes here. Would you like to take a look?"

[0238] Step 10:

[0239] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0240] Step 11:

[0241] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0242] Step 12:

[0243] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0244] Step 13:

[0245] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0246] Through the processing flow described above, the system of the present invention, including the emotion engine, enables natural interaction and guidance with the user within the metaverse, thereby improving the user experience.

[0247] (Example 2)

[0248] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0249] Traditional metaverse environments have a problem with dialogue systems with NPCs (non-player characters) that degrade the quality of the user experience because they proceed without considering the user's emotional state. Furthermore, it has been difficult to build a system that analyzes user input in real time and generates appropriate responses.

[0250] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0251] In this invention, the server includes means for receiving user input, means for analyzing the input data, means for using a generative artificial intelligence system to generate a response corresponding to the analysis results and the user's emotional state, means for transmitting the generated response to an NPC, and means for the NPC to respond to the user. This makes it possible to consider the user's emotional state and generate and provide an appropriate response in real time.

[0252] "Means of receiving user input" refers to devices or functions that capture and collect user input, whether in voice or text, within a metaverse environment.

[0253] "Means for analyzing input data" refers to software or algorithms that understand and classify the content and intent of user input data received from that user.

[0254] "Means of using generative artificial intelligence" refers to devices or functions that use artificial intelligence technology to generate appropriate responses in real time based on analyzed input data.

[0255] "Means for transmitting generated responses to NPCs" refers to devices or functions for transmitting responses generated by generative artificial intelligence to NPCs in an appropriate format.

[0256] "Means by which NPCs respond to users" refers to devices or functions that enable NPCs to interactively respond to users through voice or actions based on the responses they have generated.

[0257] "A speech recognition method that converts user voice input into text" refers to software or algorithms that convert the voice spoken by a user into digital text data.

[0258] A "natural language processing engine" refers to the technology and programs used to analyze text data and understand its content and meaning.

[0259] An "emotion engine" is software or algorithms that analyze and determine a user's emotional state from their voice and text data.

[0260] This invention is a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining generative artificial intelligence and an emotion engine for NPCs (non-player characters). This system uses the following hardware and software to analyze user input in real time and generate natural responses according to the user's emotional state.

[0261] First, the user approaches an NPC in the metaverse and speaks to them using voice or text. For example, they might say, "I'm looking for new shoes." This input is captured by the user's device (computer, smartphone, VR device, etc.). In the case of voice input, the microphone in the device collects the voice data.

[0262] The device converts the captured audio data into text data using speech recognition technology (for example, Google® Speech-to-Text API). This process generates text data such as "I'm looking for new shoes." This text data is then sent to the server.

[0263] The server uses a natural language processing (NLP) engine (for example, the Google Cloud Natural Language API) to analyze the received text data. This engine understands the user's intent and question from the text data and generates analysis results such as "looking for shoes."

[0264] Furthermore, the server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's voice or text data and recognize the user's emotional state. For example, it can determine if the user is "excited." Based on this analysis, it sends a prompt to a generative artificial intelligence (e.g., GPT-3) to generate an appropriate response. An example of a prompt might be, "Generate an appropriate response if the user is excited and looking for new shoes."

[0265] The server generates an action script for the NPC based on the response returned by the generative artificial intelligence. This script includes the content of the response to be output as voice and actions such as pointing to a specific product shelf. The server then sends this action script to the NPC.

[0266] The NPC responds to the user according to the action script it receives. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf. It can also adjust its facial expressions and tone of voice according to the user's emotions.

[0267] As a concrete example, let's look at a conversation between a user and an NPC in a fashion shop within the metaverse. When the user speaks aloud, "I'm looking for new shoes," the user's device converts the speech to text and sends it to the server. The server analyzes the received text data and understands that the user is "looking for shoes." Furthermore, if the emotion engine determines that the user is "excited," it instructs GPT-3 to "respond to an excited user" based on this analysis result. The generated response is sent to the NPC, who responds with voice and actions saying, "We have new shoes here. Would you like to take a look?"

[0268] This system can significantly improve the user experience within the metaverse. By using an emotion engine in conjunction with it, NPCs will respond in a way that takes the user's emotional state into account, resulting in more natural and satisfying conversations.

[0269] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0270] Step 1:

[0271] The user speaks to an NPC in the metaverse and says, "I'm looking for new shoes."

[0272] Input: User voice or text input.

[0273] Specific action: The user approaches an NPC through a VR device in the metaverse and speaks to them using voice, saying, "I'm looking for new shoes."

[0274] Step 2:

[0275] The audio data captured by the device is converted into text data using a speech recognition means (e.g., a speech recognition API).

[0276] Input: Captured audio data.

[0277] Output: Converted text data.

[0278] Specific operation: The user's device uses the microphone to recognize the voice and converts it into text data "looking for new shoes" through the speech recognition API. Then, this text data is sent to the server via the Internet.

[0279] Step 3:

[0280] To analyze the text data received by the server, a natural language processing engine (e.g., natural language processing API) is used.

[0281] Input: Text data sent from the user.

[0282] Output: Identification of the intention and content by analysis (e.g., "looking for shoes").

[0283] Specific operation: The server inputs the text data into the natural language processing engine and tags and analyzes the intention of "looking for shoes".

[0284] Step 4:

[0285] The server uses an emotion engine (e.g., emotion analysis API) to analyze the user's text data and recognize the emotional state.

[0286] Input: The user's text data.

[0287] Output: Emotion analysis result (e.g., "excited").

[0288] Specific operation: The server inputs the text data into the emotion engine and determines the emotion of "excited".

[0289] Step 5:

[0290] The server sends prompts to a generative AI model (e.g., a generative AI model) based on the analysis results and emotional state, and generates an appropriate response.

[0291] Input: Analysis results and emotional state.

[0292] Output: The generated response (for example, "We have new shoes here. Would you like to take a look?").

[0293] Specific operation: The server sends a prompt to the generative AI model saying, "Respond to a user looking for new shoes in an excited manner," and receives the generated response.

[0294] Step 6:

[0295] The server converts the generated response into an action script for the NPC and sends it to the NPC.

[0296] Input: Generated response.

[0297] Output: NPC action script.

[0298] Specific operation: The server converts the response into text-to-speech, generates a script that includes the actions and gestures the NPC should perform, and sends it to the NPC.

[0299] Step 7:

[0300] The NPC responds to the user based on the action script received from the server.

[0301] Input: NPC action script.

[0302] Output: Responses to the user (voice and gestures).

[0303] Specific action: The NPC will say in voice, "We have new shoes here. Would you like to take a look?" while pointing to a product shelf.

[0304] (Application Example 2)

[0305] Next, Application Example 2 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0306] In the conventional metaverse environment, non-player characters (NPCs) have limited interaction with users and can only respond based on pre-set scenarios, so there are limitations in the user experience. Also, there is no function to recognize the user's emotional state and respond accordingly, making it difficult to achieve deep conversations. As a result, there is a problem that the interaction with NPCs is unnatural and user engagement decreases.

[0307] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving the user's input, means for analyzing the input data, means for using a generative artificial intelligence that generates a response based on the analysis result, means for transmitting the generated response to an autonomous agent, means for the autonomous agent to respond to the user, means for recognizing the user's emotional state from the input data using an emotion engine, and means for generating a response according to the user's emotional state. Thereby, it becomes possible to give an appropriate and emotion-considerate response to the user's input.

[0308] The "means for receiving the user's input" is an interface for receiving an input based on voice or text that the user makes in the metaverse environment.

[0309] The "means for analyzing the input data" is a technology for analyzing the data received from the user to understand its intention and content.

[0310] The "means for using a generative artificial intelligence that generates a response" is an artificial intelligence technology for automatically generating an appropriate response based on the analyzed data.

[0311] "Means for sending to an autonomous agent" refers to a mechanism for sending the generated response to an agent that acts autonomously.

[0312] "A means by which an autonomous agent responds to a user" refers to a method by which an autonomous agent responds to a user based on the content of the response it receives.

[0313] "A means of recognizing a user's emotional state from input data using an emotion engine" refers to an engine that analyzes a user's voice or text to recognize their emotions.

[0314] "Means for generating responses that correspond to the user's emotional state" refers to technologies for generating the optimal response based on recognized emotions.

[0315] This invention is a system for making user interaction in a metaverse environment more realistic and profound, and includes the following configuration and processing steps.

[0316] System Overview

[0317] User actions

[0318] Users wear smart glasses within the metaverse and communicate with NPCs via voice and text. For example, a user might say, "I'm looking for new shoes." This input is captured by the user's device (smart glasses).

[0319] Operation of the device (smart glasses)

[0320] The smart glasses convert captured audio data into text data using a speech recognition API. Google Cloud Speech-to-Text is used as this speech recognition API. This converted text data is then sent to the server.

[0321] Server Operations

[0322] The server analyzes text data received from the terminal using a natural language processing engine (Google Cloud Natural Language). This analysis clarifies the user's intent. It also analyzes emotions from the user's voice and text using an emotion engine. This emotion engine utilizes an emotion analysis API such as IBM Watson®. Based on these analysis results, an appropriate response is generated using a generative AI model (OpenAI GPT-4®).

[0323] Controlling Autonomous Agents (NPCs)

[0324] The generated response is sent from the server to the autonomous agent. The autonomous agent provides an appropriate response to the user through voice and actions. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[0325] Specific example

[0326] 1. User: Puts on smart glasses and enters a virtual store in the metaverse. Says, "I'm looking for new shoes."

[0327] 2. Device (smart glasses): Captures the user's voice, converts it into text data such as "I'm looking for new shoes" using Google Cloud Speech-to-Text, and sends it to the server.

[0328] 3. Server: Receives text data and parses it using Google Cloud Natural Language. Understands the user's intent and analyzes the user's emotional state using an emotion engine such as IBM Watson. For example, it might recognize that the user is "excited." Based on this, it uses OpenAI GPT-4 to generate a response such as, "We have new shoes here. Would you like to take a look?"

[0329] 4. Autonomous Agent (NPC): Based on the generated response, it will approach the user and say, "We have new shoes here. Would you like to take a look?" and point to the product shelf.

[0330] Example of a prompt

[0331] Prompt text to input to the generative AI model:

[0332] The user says, "I'm looking for new shoes." The user is excited. Generate an appropriate response.

[0333] This invention provides appropriate and emotionally sensitive responses to user input, improving the user experience within the metaverse.

[0334] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0335] Processing flow of the system program that implements the application example

[0336] Step 1:

[0337] The user wears smart glasses and uses voice input within the metaverse. For example, they might say, "I'm looking for new shoes." This voice data is captured by the microphone built into the smart glasses.

[0338] Input: User's voice data

[0339] Output: Audio data

[0340] Step 2:

[0341] The device (smart glasses) converts the captured audio data into text data using a speech recognition API (Google Cloud Speech-to-Text).

[0342] Input: Audio data

[0343] Data processing: Convert audio data to text data using the Google Cloud Speech-to-Text API.

[0344] Output: Text data

[0345] Step 3:

[0346] The terminal sends the converted text data to the server.

[0347] Input: Text data

[0348] Output: Text data sent to the server

[0349] Step 4:

[0350] The server uses a natural language processing engine (Google Cloud Natural Language) to analyze text data and understand the user's intent.

[0351] Input: Text data

[0352] Data processing: Analyze user intent using Google Cloud Natural Language.

[0353] Output: User intent

[0354] Step 5:

[0355] The server uses an emotion engine (such as IBM Watson) to analyze the user's emotional state. This analysis identifies emotions from text data.

[0356] Input: Text data

[0357] Data processing: Analyze user emotions using an emotion engine.

[0358] Output: User's emotional state (e.g., excited)

[0359] Step 6:

[0360] The server uses a generative AI model (OpenAI GPT-4) to generate an appropriate response based on the analyzed user intent and emotional state. For example, it can take the prompt "The user said, 'I'm looking for new shoes.' The user is in an excited state. Generate an appropriate response." and generate a response.

[0361] Input: User intent, emotional state, prompt text

[0362] Data processing: Generate responses using OpenAI GPT-4.

[0363] Output: Generated response (Example: "We have new shoes here. Would you like to take a look?")

[0364] Step 7:

[0365] The server converts the generated response into an NPC action script and sends that script to the autonomous agent (NPC).

[0366] Input: Generated response

[0367] Data processing: Convert response text into action scripts

[0368] Output: Action script for NPCs

[0369] Step 8:

[0370] The autonomous agent (NPC) responds to the user with voice and actions according to an action script. Specifically, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf.

[0371] Input: Action script

[0372] Specific actions: The customer responds with a voice, pointing to a shelf and saying, "We have new shoes here. Would you like to take a look?"

[0373] Output: Response to the user

[0374] In this way, appropriate and emotionally sensitive responses based on user input are provided at every step.

[0375] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0376] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0377] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0378] [Second Embodiment]

[0379] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0380] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0381] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0382] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0383] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0384] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0385] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0386] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0387] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0388] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0389] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0390] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0391] This invention relates to a system that improves the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment.

[0392] Overall system configuration

[0393] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. The following describes each part of the system, subject by subject.

[0394] User-side operations

[0395] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0396] Operation on the device side

[0397] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0398] Server-side operations

[0399] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0400] The generative AI generates appropriate text responses in response to user input and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0401] NPC-side operations

[0402] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0403] Specific example

[0404] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0405] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0406] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0407] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I'm looking for shoes." Based on the analysis results, it instructs a generative AI to generate a response such as, "We have new shoes here. Would you like to take a look?"

[0408] 4. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0409] 5. NPC: Says to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[0410] 6. User: Follow the NPC's instructions and go to check out the new shoes.

[0411] In this way, the present invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, the user can achieve their goals more intuitively and effectively.

[0412] The following describes the processing flow.

[0413] Step 1:

[0414] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0415] Step 2:

[0416] The device captures the user's voice or text input. If voice input is received, the voice data is collected and prepared for processing.

[0417] Step 3:

[0418] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0419] Step 4:

[0420] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0421] Step 5:

[0422] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0423] Step 6:

[0424] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0425] Step 7:

[0426] The server sends a response generation request to the generative AI based on the analysis results of the NLP engine. The request is created based on the contextual information derived from the analysis results.

[0427] Step 8:

[0428] The generative AI receives a request from the server and generates an appropriate response. The generated response is returned to the server in text format.

[0429] Step 9:

[0430] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0431] Step 10:

[0432] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0433] Step 11:

[0434] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0435] Step 12:

[0436] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0437] Through the processing flow described above, the system of the present invention enables natural interaction and guidance with the user within the metaverse.

[0438] (Example 1)

[0439] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0440] In traditional metaverse environments, interactions with NPCs (non-player characters) were based on pre-programmed scripts, which presented challenges in adequately responding to the diverse questions and requests users might make. Furthermore, natural, real-time dialogue was difficult, limiting the user experience. As a result, users lacked flexible guidance for their actions within the metaverse, leading to reduced usability.

[0441] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0442] In this invention, the server includes means for acquiring user voice or text input, speech recognition means for converting voice input into text data, means for using a natural language processing engine to analyze the acquired text data, means for using generative artificial intelligence to generate an appropriate response based on the analysis results, means for converting the generated response into an NPC action script and sending it to the NPC, and means for the NPC to respond to the user. This makes it possible to respond flexibly to a variety of user questions and requests in real time, realize more natural and intuitive dialogue within the metaverse, and improve the user experience.

[0443] "Means for acquiring user voice or text input" refers to means for recognizing voice or text input made by a user within the metaverse and capturing it for processing within the system.

[0444] A "speech recognition means for converting voice input into text data" is a means that uses speech recognition technology to analyze a user's voice input and convert it into corresponding text data.

[0445] "Methods using a natural language processing engine to analyze acquired text data" refers to methods that utilize natural language processing technology to analyze generated text data and understand the user's intent and the content of the question.

[0446] "Means of using generative artificial intelligence to generate appropriate responses based on analysis results" refers to means of using generative artificial intelligence technology to generate appropriate responses for the user based on analysis results obtained from a natural language processing engine.

[0447] "Means for converting generated responses into action scripts for NPCs and sending them to NPCs" refers to a means of converting responses generated by a generative artificial intelligence into action scripts that can be executed by NPCs and sending them to NPCs.

[0448] "Means by which NPCs respond to users" refers to means by which NPCs provide responses to users by performing voice output or gestures based on behavioral scripts.

[0449] This invention relates to a system that enhances the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment. The following describes in detail the specific forms for implementing this system.

[0450] Overall system configuration

[0451] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and provides them to the user through NPCs. The system's components are as follows:

[0452] 1. User-side operations

[0453] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0454] 2. Operation on the device side

[0455] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0456] 3. Server-side operations

[0457] The server uses a natural language processing (NLP) engine to analyze the text data received from the terminal. The NLP engine (e.g., spaCy or NLTK) understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model, such as OpenAI's GPT-3) to generate an appropriate response. The generative AI generates an appropriate response in text for the user's input and returns it to the server. The server converts this response text into an action script for the NPC and sends it to the NPC.

[0458] 4. Actions taken by the NPC

[0459] NPCs respond to users according to behavioral scripts received from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed (e.g., pointing to a product shelf).

[0460] Specific example

[0461] Below is an example of a specific user-NPC interaction in a fashion shop within the metaverse.

[0462] 1. The user enters a fashion shop in the metaverse and speaks to an NPC saying, "I'm looking for new shoes."

[0463] 2. The device captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0464] 3. The server analyzes the received text data using an NLP engine to understand the user's intent, "I'm looking for shoes." Based on this analysis, it sends a prompt message to the AI ​​model: "The user is looking for new shoes. Please guide them to our products as an appropriate response." This prompt generates the appropriate response: "We have new shoes here. Would you like to take a look?"

[0465] 4. The server converts the generated response into an NPC action script and sends it to the NPC. The action script includes actions such as "talk" and "point to the product shelf".

[0466] 5. The NPC responds to the user with a voice message saying, "We have new shoes here. Would you like to take a look?" and then points to them.

[0467] 6. The user follows the NPC's instructions and goes to check out the new shoes.

[0468] In this way, this invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, users can achieve their goals more intuitively and effectively.

[0469] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0470] Step 1:

[0471] The user speaks to an NPC in the metaverse via voice or text. For example, the user might say, "I'm looking for new shoes." The input is the user's voice or text, and the output is captured on the user's device.

[0472] Step 2:

[0473] The device converts the captured audio data into text data using speech recognition. This process uses a speech recognition API (e.g., a speech recognition service) to convert the audio "I'm looking for new shoes" into text data with the same meaning. The text data is generated as output and passed to the next step.

[0474] Step 3:

[0475] The terminal sends the converted text data to the server. This communication uses a secure method such as the HTTPS protocol. The input is the text data generated in step 2, and the output is the text data transferred to the server.

[0476] Step 4:

[0477] The server uses a natural language processing (NLP) engine (e.g., spaCy or NLTK) to parse the received text data. Specifically, it parses the text "I'm looking for new shoes" and understands that the user's intent is "I'm looking for shoes." This parsing process is a data processing step that breaks down text data into meaning and generates the result. The input is text data, and the output is the parsed result regarding the user's intent.

[0478] Step 5:

[0479] The server sends a prompt to a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. As an example of a prompt, it generates the sentence "The user is looking for new shoes. Please guide them to products as an appropriate response" and sends it to the AI ​​model. The input is a prompt based on the analysis results, and the output is a request to the generative AI model.

[0480] Step 6:

[0481] The generative AI model generates an appropriate response based on the given prompt. For example, it generates the text response, "We have new shoes here. Would you like to take a look?" The input is the prompt, and the output is the generated response text.

[0482] Step 7:

[0483] The server converts the generated response text into an NPC action script. This action script includes actions such as speaking the response and specific gestures (e.g., pointing to a product shelf). The input is the generated response text, and the output is the action script.

[0484] Step 8:

[0485] The server sends an action script to the NPC. This prepares the NPC to provide a response to the user. The input is the action script, and the output is the script sent to the NPC.

[0486] Step 9:

[0487] The NPC responds to the user according to the action script received from the server. Using a voice output device, it might say, "We have new shoes here. Would you like to take a look?" and simultaneously point to them. The input is the action script, and the output is the response and action to the user.

[0488] (Application Example 1)

[0489] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0490] In metaverse environments, user-NPC dialogue systems are limited to standard text-based responses and simple preset responses, resulting in a restricted user experience. Virtual stores, in particular, require a concrete dialogue system that allows users to easily obtain product information and enjoy intuitive shopping. Conventional systems lack real-time natural language response, speech recognition, and speech output capabilities, often leading to a lower quality user experience.

[0491] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0492] In this invention, the server includes means for receiving user input, means for analyzing input data, means for using generative artificial intelligence to generate a response based on the analysis results, means for transmitting the generated response to a display character, means for the display character to respond to the user, and means for voice recognition and voice synthesis for voice input and voice output. This enables users visiting a virtual store to engage in natural and intuitive real-time conversations with an NPC using a head-mounted display, and to smoothly acquire product information and proceed with purchase procedures.

[0493] "User input" refers to actions and information provided by users in the metaverse environment via voice or text.

[0494] "Input data" refers to audio and text data obtained from user input.

[0495] "Analysis" is the process of understanding input data and interpreting the user's intent and the content of their questions.

[0496] "Generative artificial intelligence" is an AI technology that generates appropriate responses based on the results of analyzing input data.

[0497] A "display character" refers to an NPC (non-player character) that responds to the user in the metaverse environment.

[0498] "Speech recognition" is a technology that converts a user's voice input into text data.

[0499] "Speech synthesis" is a technology that converts text data into speech data and outputs it as speech.

[0500] "Natural language processing" refers to a set of computational techniques and methods for understanding, analyzing, and generating human language.

[0501] This invention provides a system that allows users to interact with NPCs in real time using a head-mounted display within a virtual store. Users can efficiently obtain product information and proceed smoothly with the purchase process. The configuration and processing of each part of the system are shown below.

[0502] User actions

[0503] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs) using their voice. For example, a user might say, "I'm looking for a new smartphone." This input is captured by a microphone built into the user's head-mounted display. The voice data is then converted into text data within the device.

[0504] Terminal operation

[0505] The terminal uses speech recognition software to convert the user's voice input into text data. The speech recognition software used is the Python `speech_recognition` library. The converted text data is sent to the server.

[0506] Server Operations

[0507] The server analyzes the received text data using a natural language processing engine (NLP engine). The NLP engine used is the nlptown / bert-base-multilingual-uncased-sentiment model. Based on the analyzed user intent and question content, it queries a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. This generated response text is then sent back to the user's terminal by the server and converted into an action script for the displayed character.

[0508] NPC Controls

[0509] The NPC displayed on the head-mounted display uses text-to-speech software to play back response text received from the server. The pyttsx3 library is used for this purpose. The NPC provides the generated response to the user as audio, and, if necessary, accompanies it with body movements.

[0510] Specific example

[0511] When a user says "I'm looking for a new smartphone" in the virtual store, the system responds as follows:

[0512] Voice input: "I'm looking for a new smartphone."

[0513] Analysis of user intent: "Prefers to view on a smartphone"

[0514] Response generation: "We have a new smartphone here. Would you like to take a look at the latest model?"

[0515] Example of a prompt

[0516] After analyzing the user's voice input as "I'm looking for a new smartphone," the following prompt is entered into the generative AI model:

[0517] "Generate appropriate responses for users looking for a new smartphone."

[0518] This provides a convenient system that allows users to intuitively obtain product information and effectively proceed with the purchase process.

[0519] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0520] Step 1:

[0521] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs). During this process, the user's voice input is captured by a microphone built into the head-mounted display.

[0522] Input: User voice input (e.g., "I'm looking for a new smartphone.")

[0523] Output: Captured audio data

[0524] Step 2:

[0525] The device converts the captured audio data into text data using speech recognition software. This speech recognition uses the Python `speech_recognition` library.

[0526] Input: Captured audio data

[0527] Data processing: Text conversion using speech recognition.

[0528] Output: Converted text data (e.g., "I'm looking for a new smartphone.")

[0529] Step 3:

[0530] The terminal sends the converted text data to the server.

[0531] Input: Converted text data

[0532] Output: Text data sent to the server

[0533] Step 4:

[0534] The server analyzes the received text data using a natural language processing engine. This engine uses the nlptown / bert-base-multilingual-uncased-sentiment model.

[0535] Input: Received text data

[0536] Data processing: Analyzing user intent using natural language processing

[0537] Output: Analyzed user intent (e.g., "I want to view this on my smartphone")

[0538] Step 5:

[0539] The server sends a prompt to a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. The prompt is "Generate an appropriate response for a user looking for a new smartphone."

[0540] Input: Analyzed user intent, prompt message

[0541] Data processing: Prompt-based response generation

[0542] Output: Generated response text (Example: "We have a new smartphone here. Would you like to take a look at the latest model?")

[0543] Step 6:

[0544] The server sends the generated response text to the user's terminal and converts it into a character script for display on the terminal.

[0545] Input: Generated response text

[0546] Output: Converted display character script

[0547] Step 7:

[0548] The terminal uses the converted script to have the displayed character (NPC) play a response using speech synthesis software (e.g., pyttsx3). During this process, speech synthesis generates audio data, which is then provided to the user through the NPC's speaker. The NPC also performs gestures and other actions as needed.

[0549] Input: Converted display character script

[0550] Data processing: Generation of voice data using speech synthesis, and operation instructions based on scripts.

[0551] Output: Voice responses and gestures to the user

[0552] Through these steps, the user experience within the virtual store is significantly improved, allowing users to intuitively obtain the necessary information and proceed with the purchase process smoothly.

[0553] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0554] This invention relates to a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining a generative artificial intelligence system and an emotion engine for NPCs (non-player characters).

[0555] Overall system configuration

[0556] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. Furthermore, by incorporating an emotion engine, it recognizes the user's emotional state and generates responses accordingly. The following describes each part of the system, subject by subject.

[0557] User-side operations

[0558] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0559] Operation on the device side

[0560] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0561] Server-side operations

[0562] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0563] Furthermore, the server uses an emotion engine to analyze the user's voice or text data and recognize the user's emotions. Based on the analysis results of the emotion engine, the generative AI adjusts its response. This process generates a response that is more appropriate to the user's state of mind.

[0564] The generative AI generates appropriate text responses corresponding to user input and the results of the emotion engine, and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0565] NPC-side operations

[0566] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0567] Specific example

[0568] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0569] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0570] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0571] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I am looking for shoes." Furthermore, it uses an emotion engine to analyze the user's voice data and recognize emotional states such as "excited" or "distressed."

[0572] 4. Server: Based on the analysis results and the emotion engine's results, it gives instructions to the generative AI to generate responses such as, "We have new shoes here. Would you like to take a look?" or "I'm here to help if you need anything."

[0573] 5. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0574] 6. NPC: The NPC will say to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf. It can also change its facial expressions and tone of voice in response to the user's emotions.

[0575] 7. User: Follow the NPC's instructions and go to check out the new shoes.

[0576] In this way, the system of the present invention improves the user experience within the metaverse. By combining it with an emotion engine, NPCs can not only respond flexibly to the user's specific questions and intentions, but also provide responses that take into account the user's emotional state.

[0577] The following describes the processing flow.

[0578] Step 1:

[0579] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0580] Step 2:

[0581] The device captures the user's voice or text input. In the case of voice input, it collects the voice data and prepares it for processing.

[0582] Step 3:

[0583] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0584] Step 4:

[0585] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0586] Step 5:

[0587] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0588] Step 6:

[0589] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0590] Step 7:

[0591] The server invokes an emotion engine to analyze the received audio or text data and recognize the user's emotions. For example, it extracts emotions from voice tone or text.

[0592] Step 8:

[0593] The server sends a request to the generative AI to generate a response based on the analysis results of the NLP engine and the emotion engine. For example, it provides context such as "the user is looking for shoes" or "the user is in trouble."

[0594] Step 9:

[0595] A generative AI receives a request from a server and generates an appropriate response. For example, it might generate a response like, "We have new shoes here. Would you like to take a look?"

[0596] Step 10:

[0597] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0598] Step 11:

[0599] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0600] Step 12:

[0601] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0602] Step 13:

[0603] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0604] Through the processing flow described above, the system of the present invention, including the emotion engine, enables natural interaction and guidance with the user within the metaverse, thereby improving the user experience.

[0605] (Example 2)

[0606] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0607] Traditional metaverse environments have a problem with dialogue systems with NPCs (non-player characters) that degrade the quality of the user experience because they proceed without considering the user's emotional state. Furthermore, it has been difficult to build a system that analyzes user input in real time and generates appropriate responses.

[0608] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0609] In this invention, the server includes means for receiving user input, means for analyzing the input data, means for using a generative artificial intelligence system to generate a response corresponding to the analysis results and the user's emotional state, means for transmitting the generated response to an NPC, and means for the NPC to respond to the user. This makes it possible to consider the user's emotional state and generate and provide an appropriate response in real time.

[0610] "Means of receiving user input" refers to devices or functions that capture and collect user input, whether in voice or text, within a metaverse environment.

[0611] "Means for analyzing input data" refers to software or algorithms that understand and classify the content and intent of user input data received from that user.

[0612] "Means of using generative artificial intelligence" refers to devices or functions that use artificial intelligence technology to generate appropriate responses in real time based on analyzed input data.

[0613] "Means for transmitting generated responses to NPCs" refers to devices or functions for transmitting responses generated by generative artificial intelligence to NPCs in an appropriate format.

[0614] "Means by which NPCs respond to users" refers to devices or functions that enable NPCs to interactively respond to users through voice or actions based on the responses they have generated.

[0615] "A speech recognition method that converts user voice input into text" refers to software or algorithms that convert the voice spoken by a user into digital text data.

[0616] A "natural language processing engine" refers to the technology and programs used to analyze text data and understand its content and meaning.

[0617] An "emotion engine" is software or algorithms that analyze and determine a user's emotional state from their voice and text data.

[0618] This invention is a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining generative artificial intelligence and an emotion engine for NPCs (non-player characters). This system uses the following hardware and software to analyze user input in real time and generate natural responses according to the user's emotional state.

[0619] First, the user approaches an NPC in the metaverse and speaks to them using voice or text. For example, they might say, "I'm looking for new shoes." This input is captured by the user's device (computer, smartphone, VR device, etc.). In the case of voice input, the microphone in the device collects the voice data.

[0620] The device converts the captured audio data into text data using speech recognition (for example, the Google Speech-to-Text API). This process generates text data such as "I'm looking for new shoes." This text data is then sent to the server.

[0621] The server uses a natural language processing (NLP) engine (for example, the Google Cloud Natural Language API) to analyze the received text data. This engine understands the user's intent and question from the text data and generates analysis results such as "looking for shoes."

[0622] Furthermore, the server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's voice or text data and recognize the user's emotional state. For example, it can determine if the user is "excited." Based on this analysis, it sends a prompt to a generative artificial intelligence (e.g., GPT-3) to generate an appropriate response. An example of a prompt might be, "Generate an appropriate response if the user is excited and looking for new shoes."

[0623] The server generates an action script for the NPC based on the response returned by the generative artificial intelligence. This script includes the content of the response to be output as voice and actions such as pointing to a specific product shelf. The server then sends this action script to the NPC.

[0624] The NPC responds to the user according to the action script it receives. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf. It can also adjust its facial expressions and tone of voice according to the user's emotions.

[0625] As a concrete example, let's look at a conversation between a user and an NPC in a fashion shop within the metaverse. When the user speaks aloud, "I'm looking for new shoes," the user's device converts the speech to text and sends it to the server. The server analyzes the received text data and understands that the user is "looking for shoes." Furthermore, if the emotion engine determines that the user is "excited," it instructs GPT-3 to "respond to an excited user" based on this analysis result. The generated response is sent to the NPC, who responds with voice and actions saying, "We have new shoes here. Would you like to take a look?"

[0626] This system can significantly improve the user experience within the metaverse. By using an emotion engine in conjunction with it, NPCs will respond in a way that takes the user's emotional state into account, resulting in more natural and satisfying conversations.

[0627] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0628] Step 1:

[0629] The user speaks to an NPC in the metaverse and says, "I'm looking for new shoes."

[0630] Input: User voice or text input.

[0631] Specific action: The user approaches an NPC through a VR device in the metaverse and speaks to them using voice, saying, "I'm looking for new shoes."

[0632] Step 2:

[0633] The audio data captured by the device is converted into text data using a speech recognition means (e.g., a speech recognition API).

[0634] Input: Captured audio data.

[0635] Output: Converted text data.

[0636] Specific operation: The user's device uses its microphone to recognize speech and converts it into text data, such as "I'm looking for new shoes," via a speech recognition API. This text data is then sent to the server over the internet.

[0637] Step 3:

[0638] The server uses a natural language processing engine (e.g., a natural language processing API) to analyze the text data it receives.

[0639] Input: Text data submitted by the user.

[0640] Output: Identification of intent and content through analysis (e.g., "Looking for shoes").

[0641] Specific operation: The server inputs text data into a natural language processing engine, tags it with the intention "looking for shoes," and analyzes it.

[0642] Step 4:

[0643] The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's text data and recognize their emotional state.

[0644] Input: User's text data.

[0645] Output: Sentiment analysis result (e.g., "excited").

[0646] Specific operation: The server inputs text data into the emotion engine and determines that the emotion is "excited".

[0647] Step 5:

[0648] The server sends prompts to a generative AI model (e.g., a generative AI model) based on the analysis results and emotional state, and generates an appropriate response.

[0649] Input: Analysis results and emotional state.

[0650] Output: The generated response (for example, "We have new shoes here. Would you like to take a look?").

[0651] Specific operation: The server sends a prompt to the generative AI model saying, "Respond to a user looking for new shoes in an excited manner," and receives the generated response.

[0652] Step 6:

[0653] The server converts the generated response into an action script for the NPC and sends it to the NPC.

[0654] Input: Generated response.

[0655] Output: NPC action script.

[0656] Specific operation: The server converts the response into text-to-speech, generates a script that includes the actions and gestures the NPC should perform, and sends it to the NPC.

[0657] Step 7:

[0658] The NPC responds to the user based on the action script received from the server.

[0659] Input: NPC action script.

[0660] Output: Responses to the user (voice and gestures).

[0661] Specific action: The NPC will say in voice, "We have new shoes here. Would you like to take a look?" while pointing to a product shelf.

[0662] (Application Example 2)

[0663] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0664] In traditional metaverse environments, non-player characters (NPCs) have limited interaction with the user and can only respond based on pre-set scenarios, thus limiting the user experience. Furthermore, they lack the ability to recognize and respond to the user's emotional state, making deep dialogue difficult. As a result, conversations with NPCs often felt unnatural, leading to decreased user engagement.

[0665] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving user input, means for analyzing input data, means for using a generative artificial intelligence system to generate a response based on the analysis results, means for transmitting the generated response to an autonomous agent, means for the autonomous agent to respond to the user, means for recognizing the user's emotional state from the input data using an emotion engine, and means for generating a response corresponding to the user's emotional state. This makes it possible to provide an appropriate and emotionally sensitive response to user input.

[0666] "Means for receiving user input" refers to an interface for receiving voice or text-based input from a user in the metaverse environment.

[0667] "Means for analyzing input data" refers to technologies that analyze data received from users to understand their intentions and content.

[0668] "Means using generative artificial intelligence to generate responses" refers to artificial intelligence technology for automatically generating appropriate responses based on analyzed data.

[0669] "Means for sending to an autonomous agent" refers to a mechanism for sending the generated response to an agent that acts autonomously.

[0670] "A means by which an autonomous agent responds to a user" refers to a method by which an autonomous agent responds to a user based on the content of the response it receives.

[0671] "A means of recognizing a user's emotional state from input data using an emotion engine" refers to an engine that analyzes a user's voice or text to recognize their emotions.

[0672] "Means for generating responses that correspond to the user's emotional state" refers to technologies for generating the optimal response based on recognized emotions.

[0673] This invention is a system for making user interaction in a metaverse environment more realistic and profound, and includes the following configuration and processing steps.

[0674] System Overview

[0675] User actions

[0676] Users wear smart glasses within the metaverse and communicate with NPCs via voice and text. For example, a user might say, "I'm looking for new shoes." This input is captured by the user's device (smart glasses).

[0677] Operation of the device (smart glasses)

[0678] The smart glasses convert captured audio data into text data using a speech recognition API. Google Cloud Speech-to-Text is used as this speech recognition API. This converted text data is then sent to the server.

[0679] Server Operations

[0680] The server analyzes text data received from the terminal using a natural language processing engine (Google Cloud Natural Language). This analysis clarifies the user's intent. It also analyzes emotions from the user's voice and text using an emotion engine. This emotion engine utilizes an emotion analysis API such as IBM Watson. Based on these analysis results, an appropriate response is generated using a generative AI model (OpenAI GPT-4).

[0681] Controlling Autonomous Agents (NPCs)

[0682] The generated response is sent from the server to the autonomous agent. The autonomous agent provides an appropriate response to the user through voice and actions. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[0683] Specific example

[0684] 1. User: Puts on smart glasses and enters a virtual store in the metaverse. Says, "I'm looking for new shoes."

[0685] 2. Device (smart glasses): Captures the user's voice, converts it into text data such as "I'm looking for new shoes" using Google Cloud Speech-to-Text, and sends it to the server.

[0686] 3. Server: Receives text data and parses it using Google Cloud Natural Language. Understands the user's intent and analyzes the user's emotional state using an emotion engine such as IBM Watson. For example, it might recognize that the user is "excited." Based on this, it uses OpenAI GPT-4 to generate a response such as, "We have new shoes here. Would you like to take a look?"

[0687] 4. Autonomous Agent (NPC): Based on the generated response, it will approach the user and say, "We have new shoes here. Would you like to take a look?" and point to the product shelf.

[0688] Example of a prompt

[0689] Prompt text to input to the generative AI model:

[0690] The user says, "I'm looking for new shoes." The user is excited. Generate an appropriate response.

[0691] This invention provides appropriate and emotionally sensitive responses to user input, improving the user experience within the metaverse.

[0692] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0693] Processing flow of the system program that implements the application example

[0694] Step 1:

[0695] The user wears smart glasses and uses voice input within the metaverse. For example, they might say, "I'm looking for new shoes." This voice data is captured by the microphone built into the smart glasses.

[0696] Input: User's voice data

[0697] Output: Audio data

[0698] Step 2:

[0699] The device (smart glasses) converts the captured audio data into text data using a speech recognition API (Google Cloud Speech-to-Text).

[0700] Input: Audio data

[0701] Data processing: Convert audio data to text data using the Google Cloud Speech-to-Text API.

[0702] Output: Text data

[0703] Step 3:

[0704] The terminal sends the converted text data to the server.

[0705] Input: Text data

[0706] Output: Text data sent to the server

[0707] Step 4:

[0708] The server uses a natural language processing engine (Google Cloud Natural Language) to analyze text data and understand the user's intent.

[0709] Input: Text data

[0710] Data processing: Analyze user intent using Google Cloud Natural Language.

[0711] Output: User intent

[0712] Step 5:

[0713] The server uses an emotion engine (such as IBM Watson) to analyze the user's emotional state. This analysis identifies emotions from text data.

[0714] Input: Text data

[0715] Data processing: Analyze user emotions using an emotion engine.

[0716] Output: User's emotional state (e.g., excited)

[0717] Step 6:

[0718] The server uses a generative AI model (OpenAI GPT-4) to generate an appropriate response based on the analyzed user intent and emotional state. For example, it can take the prompt "The user said, 'I'm looking for new shoes.' The user is in an excited state. Generate an appropriate response." and generate a response.

[0719] Input: User intent, emotional state, prompt text

[0720] Data processing: Generate responses using OpenAI GPT-4.

[0721] Output: Generated response (Example: "We have new shoes here. Would you like to take a look?")

[0722] Step 7:

[0723] The server converts the generated response into an NPC action script and sends that script to the autonomous agent (NPC).

[0724] Input: Generated response

[0725] Data processing: Convert response text into action scripts

[0726] Output: Action script for NPCs

[0727] Step 8:

[0728] The autonomous agent (NPC) responds to the user with voice and actions according to an action script. Specifically, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf.

[0729] Input: Action script

[0730] Specific actions: The customer responds with a voice, pointing to a shelf and saying, "We have new shoes here. Would you like to take a look?"

[0731] Output: Response to the user

[0732] In this way, appropriate and emotionally sensitive responses based on user input are provided at every step.

[0733] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0734] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0735] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0736] [Third Embodiment]

[0737] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0738] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0739] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0740] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0741] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0742] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0743] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0744] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0745] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0746] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0747] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0748] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0749] This invention relates to a system that improves the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment.

[0750] Overall system configuration

[0751] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. The following describes each part of the system, subject by subject.

[0752] User-side operations

[0753] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0754] Operation on the device side

[0755] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0756] Server-side operations

[0757] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0758] The generative AI generates appropriate text responses in response to user input and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0759] NPC-side operations

[0760] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0761] Specific example

[0762] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0763] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0764] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0765] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I'm looking for shoes." Based on the analysis results, it instructs a generative AI to generate a response such as, "We have new shoes here. Would you like to take a look?"

[0766] 4. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0767] 5. NPC: Says to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[0768] 6. User: Follow the NPC's instructions and go to check out the new shoes.

[0769] In this way, the present invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, the user can achieve their goals more intuitively and effectively.

[0770] The following describes the processing flow.

[0771] Step 1:

[0772] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0773] Step 2:

[0774] The device captures the user's voice or text input. If voice input is received, the voice data is collected and prepared for processing.

[0775] Step 3:

[0776] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0777] Step 4:

[0778] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0779] Step 5:

[0780] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0781] Step 6:

[0782] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0783] Step 7:

[0784] The server sends a response generation request to the generative AI based on the analysis results of the NLP engine. The request is created based on the contextual information derived from the analysis results.

[0785] Step 8:

[0786] The generative AI receives a request from the server and generates an appropriate response. The generated response is returned to the server in text format.

[0787] Step 9:

[0788] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0789] Step 10:

[0790] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0791] Step 11:

[0792] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0793] Step 12:

[0794] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0795] Through the processing flow described above, the system of the present invention enables natural interaction and guidance with the user within the metaverse.

[0796] (Example 1)

[0797] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0798] In traditional metaverse environments, interactions with NPCs (non-player characters) were based on pre-programmed scripts, which presented challenges in adequately responding to the diverse questions and requests users might make. Furthermore, natural, real-time dialogue was difficult, limiting the user experience. As a result, users lacked flexible guidance for their actions within the metaverse, leading to reduced usability.

[0799] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0800] In this invention, the server includes means for acquiring user voice or text input, speech recognition means for converting voice input into text data, means for using a natural language processing engine to analyze the acquired text data, means for using generative artificial intelligence to generate an appropriate response based on the analysis results, means for converting the generated response into an NPC action script and sending it to the NPC, and means for the NPC to respond to the user. This makes it possible to respond flexibly to a variety of user questions and requests in real time, realize more natural and intuitive dialogue within the metaverse, and improve the user experience.

[0801] "Means for acquiring user voice or text input" refers to means for recognizing voice or text input made by a user within the metaverse and capturing it for processing within the system.

[0802] A "speech recognition means for converting voice input into text data" is a means that uses speech recognition technology to analyze a user's voice input and convert it into corresponding text data.

[0803] "Methods using a natural language processing engine to analyze acquired text data" refers to methods that utilize natural language processing technology to analyze generated text data and understand the user's intent and the content of the question.

[0804] "Means of using generative artificial intelligence to generate appropriate responses based on analysis results" refers to means of using generative artificial intelligence technology to generate appropriate responses for the user based on analysis results obtained from a natural language processing engine.

[0805] "Means for converting generated responses into action scripts for NPCs and sending them to NPCs" refers to a means of converting responses generated by a generative artificial intelligence into action scripts that can be executed by NPCs and sending them to NPCs.

[0806] "Means by which NPCs respond to users" refers to means by which NPCs provide responses to users by performing voice output or gestures based on behavioral scripts.

[0807] This invention relates to a system that enhances the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment. The following describes in detail the specific forms for implementing this system.

[0808] Overall system configuration

[0809] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and provides them to the user through NPCs. The system's components are as follows:

[0810] 1. User-side operations

[0811] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0812] 2. Operation on the device side

[0813] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0814] 3. Server-side operations

[0815] The server uses a natural language processing (NLP) engine to analyze the text data received from the terminal. The NLP engine (e.g., spaCy or NLTK) understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model, such as OpenAI's GPT-3) to generate an appropriate response. The generative AI generates an appropriate response in text for the user's input and returns it to the server. The server converts this response text into an action script for the NPC and sends it to the NPC.

[0816] 4. Actions taken by the NPC

[0817] NPCs respond to users according to behavioral scripts received from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed (e.g., pointing to a product shelf).

[0818] Specific example

[0819] Below is an example of a specific user-NPC interaction in a fashion shop within the metaverse.

[0820] 1. The user enters a fashion shop in the metaverse and speaks to an NPC saying, "I'm looking for new shoes."

[0821] 2. The device captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0822] 3. The server analyzes the received text data using an NLP engine to understand the user's intent, "I'm looking for shoes." Based on this analysis, it sends a prompt message to the AI ​​model: "The user is looking for new shoes. Please guide them to our products as an appropriate response." This prompt generates the appropriate response: "We have new shoes here. Would you like to take a look?"

[0823] 4. The server converts the generated response into an NPC action script and sends it to the NPC. The action script includes actions such as "talk" and "point to the product shelf".

[0824] 5. The NPC responds to the user with a voice message saying, "We have new shoes here. Would you like to take a look?" and then points to them.

[0825] 6. The user follows the NPC's instructions and goes to check out the new shoes.

[0826] In this way, this invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, users can achieve their goals more intuitively and effectively.

[0827] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0828] Step 1:

[0829] The user speaks to an NPC in the metaverse via voice or text. For example, the user might say, "I'm looking for new shoes." The input is the user's voice or text, and the output is captured on the user's device.

[0830] Step 2:

[0831] The device converts the captured audio data into text data using speech recognition. This process uses a speech recognition API (e.g., a speech recognition service) to convert the audio "I'm looking for new shoes" into text data with the same meaning. The text data is generated as output and passed to the next step.

[0832] Step 3:

[0833] The terminal sends the converted text data to the server. This communication uses a secure method such as the HTTPS protocol. The input is the text data generated in step 2, and the output is the text data transferred to the server.

[0834] Step 4:

[0835] The server uses a natural language processing (NLP) engine (e.g., spaCy or NLTK) to parse the received text data. Specifically, it parses the text "I'm looking for new shoes" and understands that the user's intent is "I'm looking for shoes." This parsing process is a data processing step that breaks down text data into meaning and generates the result. The input is text data, and the output is the parsed result regarding the user's intent.

[0836] Step 5:

[0837] The server sends a prompt to a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. As an example of a prompt, it generates the sentence "The user is looking for new shoes. Please guide them to products as an appropriate response" and sends it to the AI ​​model. The input is a prompt based on the analysis results, and the output is a request to the generative AI model.

[0838] Step 6:

[0839] The generative AI model generates an appropriate response based on the given prompt. For example, it generates the text response, "We have new shoes here. Would you like to take a look?" The input is the prompt, and the output is the generated response text.

[0840] Step 7:

[0841] The server converts the generated response text into an NPC action script. This action script includes actions such as speaking the response and specific gestures (e.g., pointing to a product shelf). The input is the generated response text, and the output is the action script.

[0842] Step 8:

[0843] The server sends an action script to the NPC. This prepares the NPC to provide a response to the user. The input is the action script, and the output is the script sent to the NPC.

[0844] Step 9:

[0845] The NPC responds to the user according to the action script received from the server. Using a voice output device, it might say, "We have new shoes here. Would you like to take a look?" and simultaneously point to them. The input is the action script, and the output is the response and action to the user.

[0846] (Application Example 1)

[0847] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0848] In metaverse environments, user-NPC dialogue systems are limited to standard text-based responses and simple preset responses, resulting in a restricted user experience. Virtual stores, in particular, require a concrete dialogue system that allows users to easily obtain product information and enjoy intuitive shopping. Conventional systems lack real-time natural language response, speech recognition, and speech output capabilities, often leading to a lower quality user experience.

[0849] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0850] In this invention, the server includes means for receiving user input, means for analyzing input data, means for using generative artificial intelligence to generate a response based on the analysis results, means for transmitting the generated response to a display character, means for the display character to respond to the user, and means for voice recognition and voice synthesis for voice input and voice output. This enables users visiting a virtual store to engage in natural and intuitive real-time conversations with an NPC using a head-mounted display, and to smoothly acquire product information and proceed with purchase procedures.

[0851] "User input" refers to actions and information provided by users in the metaverse environment via voice or text.

[0852] "Input data" refers to audio and text data obtained from user input.

[0853] "Analysis" is the process of understanding input data and interpreting the user's intent and the content of their questions.

[0854] "Generative artificial intelligence" is an AI technology that generates appropriate responses based on the results of analyzing input data.

[0855] A "display character" refers to an NPC (non-player character) that responds to the user in the metaverse environment.

[0856] "Speech recognition" is a technology that converts a user's voice input into text data.

[0857] "Speech synthesis" is a technology that converts text data into speech data and outputs it as speech.

[0858] "Natural language processing" refers to a set of computational techniques and methods for understanding, analyzing, and generating human language.

[0859] This invention provides a system that allows users to interact with NPCs in real time using a head-mounted display within a virtual store. Users can efficiently obtain product information and proceed smoothly with the purchase process. The configuration and processing of each part of the system are shown below.

[0860] User actions

[0861] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs) using their voice. For example, a user might say, "I'm looking for a new smartphone." This input is captured by a microphone built into the user's head-mounted display. The voice data is then converted into text data within the device.

[0862] Terminal operation

[0863] The terminal uses speech recognition software to convert the user's voice input into text data. The speech recognition software used is the Python `speech_recognition` library. The converted text data is sent to the server.

[0864] Server Operations

[0865] The server analyzes the received text data using a natural language processing engine (NLP engine). The NLP engine used is the nlptown / bert-base-multilingual-uncased-sentiment model. Based on the analyzed user intent and question content, it queries a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. This generated response text is then sent back to the user's terminal by the server and converted into an action script for the displayed character.

[0866] NPC Controls

[0867] The NPC displayed on the head-mounted display uses text-to-speech software to play back response text received from the server. The pyttsx3 library is used for this purpose. The NPC provides the generated response to the user as audio, and, if necessary, accompanies it with body movements.

[0868] Specific example

[0869] When a user says "I'm looking for a new smartphone" in the virtual store, the system responds as follows:

[0870] Voice input: "I'm looking for a new smartphone."

[0871] Analysis of user intent: "Prefers to view on a smartphone"

[0872] Response generation: "We have a new smartphone here. Would you like to take a look at the latest model?"

[0873] Example of a prompt

[0874] After analyzing the user's voice input as "I'm looking for a new smartphone," the following prompt is entered into the generative AI model:

[0875] "Generate appropriate responses for users looking for a new smartphone."

[0876] This provides a convenient system that allows users to intuitively obtain product information and effectively proceed with the purchase process.

[0877] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0878] Step 1:

[0879] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs). During this process, the user's voice input is captured by a microphone built into the head-mounted display.

[0880] Input: User voice input (e.g., "I'm looking for a new smartphone.")

[0881] Output: Captured audio data

[0882] Step 2:

[0883] The device converts the captured audio data into text data using speech recognition software. This speech recognition uses the Python `speech_recognition` library.

[0884] Input: Captured audio data

[0885] Data processing: Text conversion using speech recognition.

[0886] Output: Converted text data (e.g., "I'm looking for a new smartphone.")

[0887] Step 3:

[0888] The terminal sends the converted text data to the server.

[0889] Input: Converted text data

[0890] Output: Text data sent to the server

[0891] Step 4:

[0892] The server analyzes the received text data using a natural language processing engine. This engine uses the nlptown / bert-base-multilingual-uncased-sentiment model.

[0893] Input: Received text data

[0894] Data processing: Analyzing user intent using natural language processing

[0895] Output: Analyzed user intent (e.g., "I want to view this on my smartphone")

[0896] Step 5:

[0897] The server sends a prompt to a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. The prompt is "Generate an appropriate response for a user looking for a new smartphone."

[0898] Input: Analyzed user intent, prompt message

[0899] Data processing: Prompt-based response generation

[0900] Output: Generated response text (Example: "We have a new smartphone here. Would you like to take a look at the latest model?")

[0901] Step 6:

[0902] The server sends the generated response text to the user's terminal and converts it into a character script for display on the terminal.

[0903] Input: Generated response text

[0904] Output: Converted display character script

[0905] Step 7:

[0906] The terminal uses the converted script to have the displayed character (NPC) play a response using speech synthesis software (e.g., pyttsx3). During this process, speech synthesis generates audio data, which is then provided to the user through the NPC's speaker. The NPC also performs gestures and other actions as needed.

[0907] Input: Converted display character script

[0908] Data processing: Generation of voice data using speech synthesis, and operation instructions based on scripts.

[0909] Output: Voice responses and gestures to the user

[0910] Through these steps, the user experience within the virtual store is significantly improved, allowing users to intuitively obtain the necessary information and proceed with the purchase process smoothly.

[0911] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0912] This invention relates to a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining a generative artificial intelligence system and an emotion engine for NPCs (non-player characters).

[0913] Overall system configuration

[0914] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. Furthermore, by incorporating an emotion engine, it recognizes the user's emotional state and generates responses accordingly. The following describes each part of the system, subject by subject.

[0915] User-side operations

[0916] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[0917] Operation on the device side

[0918] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[0919] Server-side operations

[0920] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[0921] Furthermore, the server uses an emotion engine to analyze the user's voice or text data and recognize the user's emotions. Based on the analysis results of the emotion engine, the generative AI adjusts its response. This process generates a response that is more appropriate to the user's state of mind.

[0922] The generative AI generates appropriate text responses corresponding to user input and the results of the emotion engine, and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[0923] NPC-side operations

[0924] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[0925] Specific example

[0926] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[0927] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[0928] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[0929] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I am looking for shoes." Furthermore, it uses an emotion engine to analyze the user's voice data and recognize emotional states such as "excited" or "distressed."

[0930] 4. Server: Based on the analysis results and the emotion engine's results, it gives instructions to the generative AI to generate responses such as, "We have new shoes here. Would you like to take a look?" or "I'm here to help if you need anything."

[0931] 5. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[0932] 6. NPC: The NPC will say to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf. It can also change its facial expressions and tone of voice in response to the user's emotions.

[0933] 7. User: Follow the NPC's instructions and go to check out the new shoes.

[0934] In this way, the system of the present invention improves the user experience within the metaverse. By combining it with an emotion engine, NPCs can not only respond flexibly to the user's specific questions and intentions, but also provide responses that take into account the user's emotional state.

[0935] The following describes the processing flow.

[0936] Step 1:

[0937] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[0938] Step 2:

[0939] The device captures the user's voice or text input. In the case of voice input, it collects the voice data and prepares it for processing.

[0940] Step 3:

[0941] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[0942] Step 4:

[0943] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[0944] Step 5:

[0945] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[0946] Step 6:

[0947] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[0948] Step 7:

[0949] The server invokes an emotion engine to analyze the received audio or text data and recognize the user's emotions. For example, it extracts emotions from voice tone or text.

[0950] Step 8:

[0951] The server sends a request to the generative AI to generate a response based on the analysis results of the NLP engine and the emotion engine. For example, it provides context such as "the user is looking for shoes" or "the user is in trouble."

[0952] Step 9:

[0953] A generative AI receives a request from a server and generates an appropriate response. For example, it might generate a response like, "We have new shoes here. Would you like to take a look?"

[0954] Step 10:

[0955] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[0956] Step 11:

[0957] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[0958] Step 12:

[0959] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[0960] Step 13:

[0961] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[0962] Through the processing flow described above, the system of the present invention, including the emotion engine, enables natural interaction and guidance with the user within the metaverse, thereby improving the user experience.

[0963] (Example 2)

[0964] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0965] Traditional metaverse environments have a problem with dialogue systems with NPCs (non-player characters) that degrade the quality of the user experience because they proceed without considering the user's emotional state. Furthermore, it has been difficult to build a system that analyzes user input in real time and generates appropriate responses.

[0966] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0967] In this invention, the server includes means for receiving user input, means for analyzing the input data, means for using a generative artificial intelligence system to generate a response corresponding to the analysis results and the user's emotional state, means for transmitting the generated response to an NPC, and means for the NPC to respond to the user. This makes it possible to consider the user's emotional state and generate and provide an appropriate response in real time.

[0968] "Means of receiving user input" refers to devices or functions that capture and collect user input, whether in voice or text, within a metaverse environment.

[0969] "Means for analyzing input data" refers to software or algorithms that understand and classify the content and intent of user input data received from that user.

[0970] "Means of using generative artificial intelligence" refers to devices or functions that use artificial intelligence technology to generate appropriate responses in real time based on analyzed input data.

[0971] "Means for transmitting generated responses to NPCs" refers to devices or functions for transmitting responses generated by generative artificial intelligence to NPCs in an appropriate format.

[0972] "Means by which NPCs respond to users" refers to devices or functions that enable NPCs to interactively respond to users through voice or actions based on the responses they have generated.

[0973] "A speech recognition method that converts user voice input into text" refers to software or algorithms that convert the voice spoken by a user into digital text data.

[0974] A "natural language processing engine" refers to the technology and programs used to analyze text data and understand its content and meaning.

[0975] An "emotion engine" is software or algorithms that analyze and determine a user's emotional state from their voice and text data.

[0976] This invention is a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining generative artificial intelligence and an emotion engine for NPCs (non-player characters). This system uses the following hardware and software to analyze user input in real time and generate natural responses according to the user's emotional state.

[0977] First, the user approaches an NPC in the metaverse and speaks to them using voice or text. For example, they might say, "I'm looking for new shoes." This input is captured by the user's device (computer, smartphone, VR device, etc.). In the case of voice input, the microphone in the device collects the voice data.

[0978] The device converts the captured audio data into text data using speech recognition (for example, the Google Speech-to-Text API). This process generates text data such as "I'm looking for new shoes." This text data is then sent to the server.

[0979] The server uses a natural language processing (NLP) engine (for example, the Google Cloud Natural Language API) to analyze the received text data. This engine understands the user's intent and question from the text data and generates analysis results such as "looking for shoes."

[0980] Furthermore, the server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's voice or text data and recognize the user's emotional state. For example, it can determine if the user is "excited." Based on this analysis, it sends a prompt to a generative artificial intelligence (e.g., GPT-3) to generate an appropriate response. An example of a prompt might be, "Generate an appropriate response if the user is excited and looking for new shoes."

[0981] The server generates an action script for the NPC based on the response returned by the generative artificial intelligence. This script includes the content of the response to be output as voice and actions such as pointing to a specific product shelf. The server then sends this action script to the NPC.

[0982] The NPC responds to the user according to the action script it receives. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf. It can also adjust its facial expressions and tone of voice according to the user's emotions.

[0983] As a concrete example, let's look at a conversation between a user and an NPC in a fashion shop within the metaverse. When the user speaks aloud, "I'm looking for new shoes," the user's device converts the speech to text and sends it to the server. The server analyzes the received text data and understands that the user is "looking for shoes." Furthermore, if the emotion engine determines that the user is "excited," it instructs GPT-3 to "respond to an excited user" based on this analysis result. The generated response is sent to the NPC, who responds with voice and actions saying, "We have new shoes here. Would you like to take a look?"

[0984] This system can significantly improve the user experience within the metaverse. By using an emotion engine in conjunction with it, NPCs will respond in a way that takes the user's emotional state into account, resulting in more natural and satisfying conversations.

[0985] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0986] Step 1:

[0987] The user speaks to an NPC in the metaverse and says, "I'm looking for new shoes."

[0988] Input: User voice or text input.

[0989] Specific action: The user approaches an NPC through a VR device in the metaverse and speaks to them using voice, saying, "I'm looking for new shoes."

[0990] Step 2:

[0991] The audio data captured by the device is converted into text data using a speech recognition means (e.g., a speech recognition API).

[0992] Input: Captured audio data.

[0993] Output: Converted text data.

[0994] Specific operation: The user's device uses its microphone to recognize speech and converts it into text data, such as "I'm looking for new shoes," via a speech recognition API. This text data is then sent to the server over the internet.

[0995] Step 3:

[0996] The server uses a natural language processing engine (e.g., a natural language processing API) to analyze the text data it receives.

[0997] Input: Text data submitted by the user.

[0998] Output: Identification of intent and content through analysis (e.g., "Looking for shoes").

[0999] Specific operation: The server inputs text data into a natural language processing engine, tags it with the intention "looking for shoes," and analyzes it.

[1000] Step 4:

[1001] The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's text data and recognize their emotional state.

[1002] Input: User's text data.

[1003] Output: Sentiment analysis result (e.g., "excited").

[1004] Specific operation: The server inputs text data into the emotion engine and determines that the emotion is "excited".

[1005] Step 5:

[1006] The server sends prompts to a generative AI model (e.g., a generative AI model) based on the analysis results and emotional state, and generates an appropriate response.

[1007] Input: Analysis results and emotional state.

[1008] Output: The generated response (for example, "We have new shoes here. Would you like to take a look?").

[1009] Specific operation: The server sends a prompt to the generative AI model saying, "Respond to a user looking for new shoes in an excited manner," and receives the generated response.

[1010] Step 6:

[1011] The server converts the generated response into an action script for the NPC and sends it to the NPC.

[1012] Input: Generated response.

[1013] Output: NPC action script.

[1014] Specific operation: The server converts the response into text-to-speech, generates a script that includes the actions and gestures the NPC should perform, and sends it to the NPC.

[1015] Step 7:

[1016] The NPC responds to the user based on the action script received from the server.

[1017] Input: NPC action script.

[1018] Output: Responses to the user (voice and gestures).

[1019] Specific action: The NPC will say in voice, "We have new shoes here. Would you like to take a look?" while pointing to a product shelf.

[1020] (Application Example 2)

[1021] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1022] In traditional metaverse environments, non-player characters (NPCs) have limited interaction with the user and can only respond based on pre-set scenarios, thus limiting the user experience. Furthermore, they lack the ability to recognize and respond to the user's emotional state, making deep dialogue difficult. As a result, conversations with NPCs often felt unnatural, leading to decreased user engagement.

[1023] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving user input, means for analyzing input data, means for using a generative artificial intelligence system to generate a response based on the analysis results, means for transmitting the generated response to an autonomous agent, means for the autonomous agent to respond to the user, means for recognizing the user's emotional state from the input data using an emotion engine, and means for generating a response corresponding to the user's emotional state. This makes it possible to provide an appropriate and emotionally sensitive response to user input.

[1024] "Means for receiving user input" refers to an interface for receiving voice or text-based input from a user in the metaverse environment.

[1025] "Means for analyzing input data" refers to technologies that analyze data received from users to understand their intentions and content.

[1026] "Means using generative artificial intelligence to generate responses" refers to artificial intelligence technology for automatically generating appropriate responses based on analyzed data.

[1027] "Means for sending to an autonomous agent" refers to a mechanism for sending the generated response to an agent that acts autonomously.

[1028] "A means by which an autonomous agent responds to a user" refers to a method by which an autonomous agent responds to a user based on the content of the response it receives.

[1029] "A means of recognizing a user's emotional state from input data using an emotion engine" refers to an engine that analyzes a user's voice or text to recognize their emotions.

[1030] "Means for generating responses that correspond to the user's emotional state" refers to technologies for generating the optimal response based on recognized emotions.

[1031] This invention is a system for making user interaction in a metaverse environment more realistic and profound, and includes the following configuration and processing steps.

[1032] System Overview

[1033] User actions

[1034] Users wear smart glasses within the metaverse and communicate with NPCs via voice and text. For example, a user might say, "I'm looking for new shoes." This input is captured by the user's device (smart glasses).

[1035] Operation of the device (smart glasses)

[1036] The smart glasses convert captured audio data into text data using a speech recognition API. Google Cloud Speech-to-Text is used as this speech recognition API. This converted text data is then sent to the server.

[1037] Server Operations

[1038] The server analyzes text data received from the terminal using a natural language processing engine (Google Cloud Natural Language). This analysis clarifies the user's intent. It also analyzes emotions from the user's voice and text using an emotion engine. This emotion engine utilizes an emotion analysis API such as IBM Watson. Based on these analysis results, an appropriate response is generated using a generative AI model (OpenAI GPT-4).

[1039] Controlling Autonomous Agents (NPCs)

[1040] The generated response is sent from the server to the autonomous agent. The autonomous agent provides an appropriate response to the user through voice and actions. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[1041] Specific example

[1042] 1. User: Puts on smart glasses and enters a virtual store in the metaverse. Says, "I'm looking for new shoes."

[1043] 2. Device (smart glasses): Captures the user's voice, converts it into text data such as "I'm looking for new shoes" using Google Cloud Speech-to-Text, and sends it to the server.

[1044] 3. Server: Receives text data and parses it using Google Cloud Natural Language. Understands the user's intent and analyzes the user's emotional state using an emotion engine such as IBM Watson. For example, it might recognize that the user is "excited." Based on this, it uses OpenAI GPT-4 to generate a response such as, "We have new shoes here. Would you like to take a look?"

[1045] 4. Autonomous Agent (NPC): Based on the generated response, it will approach the user and say, "We have new shoes here. Would you like to take a look?" and point to the product shelf.

[1046] Example of a prompt

[1047] Prompt text to input to the generative AI model:

[1048] The user says, "I'm looking for new shoes." The user is excited. Generate an appropriate response.

[1049] This invention provides appropriate and emotionally sensitive responses to user input, improving the user experience within the metaverse.

[1050] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1051] Processing flow of the system program that implements the application example

[1052] Step 1:

[1053] The user wears smart glasses and uses voice input within the metaverse. For example, they might say, "I'm looking for new shoes." This voice data is captured by the microphone built into the smart glasses.

[1054] Input: User's voice data

[1055] Output: Audio data

[1056] Step 2:

[1057] The device (smart glasses) converts the captured audio data into text data using a speech recognition API (Google Cloud Speech-to-Text).

[1058] Input: Audio data

[1059] Data processing: Convert audio data to text data using the Google Cloud Speech-to-Text API.

[1060] Output: Text data

[1061] Step 3:

[1062] The terminal sends the converted text data to the server.

[1063] Input: Text data

[1064] Output: Text data sent to the server

[1065] Step 4:

[1066] The server uses a natural language processing engine (Google Cloud Natural Language) to analyze text data and understand the user's intent.

[1067] Input: Text data

[1068] Data processing: Analyze user intent using Google Cloud Natural Language.

[1069] Output: User intent

[1070] Step 5:

[1071] The server uses an emotion engine (such as IBM Watson) to analyze the user's emotional state. This analysis identifies emotions from text data.

[1072] Input: Text data

[1073] Data processing: Analyze user emotions using an emotion engine.

[1074] Output: User's emotional state (e.g., excited)

[1075] Step 6:

[1076] The server uses a generative AI model (OpenAI GPT-4) to generate an appropriate response based on the analyzed user intent and emotional state. For example, it can take the prompt "The user said, 'I'm looking for new shoes.' The user is in an excited state. Generate an appropriate response." and generate a response.

[1077] Input: User intent, emotional state, prompt text

[1078] Data processing: Generate responses using OpenAI GPT-4.

[1079] Output: Generated response (Example: "We have new shoes here. Would you like to take a look?")

[1080] Step 7:

[1081] The server converts the generated response into an NPC action script and sends that script to the autonomous agent (NPC).

[1082] Input: Generated response

[1083] Data processing: Convert response text into action scripts

[1084] Output: Action script for NPCs

[1085] Step 8:

[1086] The autonomous agent (NPC) responds to the user with voice and actions according to an action script. Specifically, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf.

[1087] Input: Action script

[1088] Specific actions: The customer responds with a voice, pointing to a shelf and saying, "We have new shoes here. Would you like to take a look?"

[1089] Output: Response to the user

[1090] In this way, appropriate and emotionally sensitive responses based on user input are provided at every step.

[1091] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1092] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1093] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1094] [Fourth Embodiment]

[1095] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1096] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1097] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1098] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1099] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1100] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1101] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1102] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1103] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1104] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1105] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1106] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1107] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1108] This invention relates to a system that improves the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment.

[1109] Overall system configuration

[1110] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. The following describes each part of the system, subject by subject.

[1111] User-side operations

[1112] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[1113] Operation on the device side

[1114] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[1115] Server-side operations

[1116] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[1117] The generative AI generates appropriate text responses in response to user input and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[1118] NPC-side operations

[1119] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[1120] Specific example

[1121] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[1122] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[1123] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[1124] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I'm looking for shoes." Based on the analysis results, it instructs a generative AI to generate a response such as, "We have new shoes here. Would you like to take a look?"

[1125] 4. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[1126] 5. NPC: Says to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[1127] 6. User: Follow the NPC's instructions and go to check out the new shoes.

[1128] In this way, the present invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, the user can achieve their goals more intuitively and effectively.

[1129] The following describes the processing flow.

[1130] Step 1:

[1131] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[1132] Step 2:

[1133] The device captures the user's voice or text input. If voice input is received, the voice data is collected and prepared for processing.

[1134] Step 3:

[1135] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[1136] Step 4:

[1137] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[1138] Step 5:

[1139] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[1140] Step 6:

[1141] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[1142] Step 7:

[1143] The server sends a response generation request to the generative AI based on the analysis results of the NLP engine. The request is created based on the contextual information derived from the analysis results.

[1144] Step 8:

[1145] The generative AI receives a request from the server and generates an appropriate response. The generated response is returned to the server in text format.

[1146] Step 9:

[1147] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[1148] Step 10:

[1149] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[1150] Step 11:

[1151] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[1152] Step 12:

[1153] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[1154] Through the processing flow described above, the system of the present invention enables natural interaction and guidance with the user within the metaverse.

[1155] (Example 1)

[1156] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1157] In traditional metaverse environments, interactions with NPCs (non-player characters) were based on pre-programmed scripts, which presented challenges in adequately responding to the diverse questions and requests users might make. Furthermore, natural, real-time dialogue was difficult, limiting the user experience. As a result, users lacked flexible guidance for their actions within the metaverse, leading to reduced usability.

[1158] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1159] In this invention, the server includes means for acquiring user voice or text input, speech recognition means for converting voice input into text data, means for using a natural language processing engine to analyze the acquired text data, means for using generative artificial intelligence to generate an appropriate response based on the analysis results, means for converting the generated response into an NPC action script and sending it to the NPC, and means for the NPC to respond to the user. This makes it possible to respond flexibly to a variety of user questions and requests in real time, realize more natural and intuitive dialogue within the metaverse, and improve the user experience.

[1160] "Means for acquiring user voice or text input" refers to means for recognizing voice or text input made by a user within the metaverse and capturing it for processing within the system.

[1161] A "speech recognition means for converting voice input into text data" is a means that uses speech recognition technology to analyze a user's voice input and convert it into corresponding text data.

[1162] "Methods using a natural language processing engine to analyze acquired text data" refers to methods that utilize natural language processing technology to analyze generated text data and understand the user's intent and the content of the question.

[1163] "Means of using generative artificial intelligence to generate appropriate responses based on analysis results" refers to means of using generative artificial intelligence technology to generate appropriate responses for the user based on analysis results obtained from a natural language processing engine.

[1164] "Means for converting generated responses into action scripts for NPCs and sending them to NPCs" refers to a means of converting responses generated by a generative artificial intelligence into action scripts that can be executed by NPCs and sending them to NPCs.

[1165] "Means by which NPCs respond to users" refers to means by which NPCs provide responses to users by performing voice output or gestures based on behavioral scripts.

[1166] This invention relates to a system that enhances the user experience by enabling real-time interaction with users through the combination of NPCs (non-player characters) and generative artificial intelligence (AI) in a metaverse environment. The following describes in detail the specific forms for implementing this system.

[1167] Overall system configuration

[1168] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and provides them to the user through NPCs. The system's components are as follows:

[1169] 1. User-side operations

[1170] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[1171] 2. Operation on the device side

[1172] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[1173] 3. Server-side operations

[1174] The server uses a natural language processing (NLP) engine to analyze the text data received from the terminal. The NLP engine (e.g., spaCy or NLTK) understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model, such as OpenAI's GPT-3) to generate an appropriate response. The generative AI generates an appropriate response in text for the user's input and returns it to the server. The server converts this response text into an action script for the NPC and sends it to the NPC.

[1175] 4. Actions taken by the NPC

[1176] NPCs respond to users according to behavioral scripts received from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed (e.g., pointing to a product shelf).

[1177] Specific example

[1178] Below is an example of a specific user-NPC interaction in a fashion shop within the metaverse.

[1179] 1. The user enters a fashion shop in the metaverse and speaks to an NPC saying, "I'm looking for new shoes."

[1180] 2. The device captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[1181] 3. The server analyzes the received text data using an NLP engine to understand the user's intent, "I'm looking for shoes." Based on this analysis, it sends a prompt message to the AI ​​model: "The user is looking for new shoes. Please guide them to our products as an appropriate response." This prompt generates the appropriate response: "We have new shoes here. Would you like to take a look?"

[1182] 4. The server converts the generated response into an NPC action script and sends it to the NPC. The action script includes actions such as "talk" and "point to the product shelf".

[1183] 5. The NPC responds to the user with a voice message saying, "We have new shoes here. Would you like to take a look?" and then points to them.

[1184] 6. The user follows the NPC's instructions and goes to check out the new shoes.

[1185] In this way, this invention improves the user experience within the metaverse. Because NPCs can respond flexibly to the user's specific questions and intentions, users can achieve their goals more intuitively and effectively.

[1186] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1187] Step 1:

[1188] The user speaks to an NPC in the metaverse via voice or text. For example, the user might say, "I'm looking for new shoes." The input is the user's voice or text, and the output is captured on the user's device.

[1189] Step 2:

[1190] The device converts the captured audio data into text data using speech recognition. This process uses a speech recognition API (e.g., a speech recognition service) to convert the audio "I'm looking for new shoes" into text data with the same meaning. The text data is generated as output and passed to the next step.

[1191] Step 3:

[1192] The terminal sends the converted text data to the server. This communication uses a secure method such as the HTTPS protocol. The input is the text data generated in step 2, and the output is the text data transferred to the server.

[1193] Step 4:

[1194] The server uses a natural language processing (NLP) engine (e.g., spaCy or NLTK) to parse the received text data. Specifically, it parses the text "I'm looking for new shoes" and understands that the user's intent is "I'm looking for shoes." This parsing process is a data processing step that breaks down text data into meaning and generates the result. The input is text data, and the output is the parsed result regarding the user's intent.

[1195] Step 5:

[1196] The server sends a prompt to a generative AI model (e.g., OpenAI's GPT-3) based on the analysis results. As an example of a prompt, it generates the sentence "The user is looking for new shoes. Please guide them to products as an appropriate response" and sends it to the AI ​​model. The input is a prompt based on the analysis results, and the output is a request to the generative AI model.

[1197] Step 6:

[1198] The generative AI model generates an appropriate response based on the given prompt. For example, it generates the text response, "We have new shoes here. Would you like to take a look?" The input is the prompt, and the output is the generated response text.

[1199] Step 7:

[1200] The server converts the generated response text into an NPC action script. This action script includes actions such as speaking the response and specific gestures (e.g., pointing to a product shelf). The input is the generated response text, and the output is the action script.

[1201] Step 8:

[1202] The server sends an action script to the NPC. This prepares the NPC to provide a response to the user. The input is the action script, and the output is the script sent to the NPC.

[1203] Step 9:

[1204] The NPC responds to the user according to the action script received from the server. Using a voice output device, it might say, "We have new shoes here. Would you like to take a look?" and simultaneously point to them. The input is the action script, and the output is the response and action to the user.

[1205] (Application Example 1)

[1206] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1207] In metaverse environments, user-NPC dialogue systems are limited to standard text-based responses and simple preset responses, resulting in a restricted user experience. Virtual stores, in particular, require a concrete dialogue system that allows users to easily obtain product information and enjoy intuitive shopping. Conventional systems lack real-time natural language response, speech recognition, and speech output capabilities, often leading to a lower quality user experience.

[1208] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1209] In this invention, the server includes means for receiving user input, means for analyzing input data, means for using generative artificial intelligence to generate a response based on the analysis results, means for transmitting the generated response to a display character, means for the display character to respond to the user, and means for voice recognition and voice synthesis for voice input and voice output. This enables users visiting a virtual store to engage in natural and intuitive real-time conversations with an NPC using a head-mounted display, and to smoothly acquire product information and proceed with purchase procedures.

[1210] "User input" refers to actions and information provided by users in the metaverse environment via voice or text.

[1211] "Input data" refers to audio and text data obtained from user input.

[1212] "Analysis" is the process of understanding input data and interpreting the user's intent and the content of their questions.

[1213] "Generative artificial intelligence" is an AI technology that generates appropriate responses based on the results of analyzing input data.

[1214] A "display character" refers to an NPC (non-player character) that responds to the user in the metaverse environment.

[1215] "Speech recognition" is a technology that converts a user's voice input into text data.

[1216] "Speech synthesis" is a technology that converts text data into speech data and outputs it as speech.

[1217] "Natural language processing" refers to a set of computational techniques and methods for understanding, analyzing, and generating human language.

[1218] This invention provides a system that allows users to interact with NPCs in real time using a head-mounted display within a virtual store. Users can efficiently obtain product information and proceed smoothly with the purchase process. The configuration and processing of each part of the system are shown below.

[1219] User actions

[1220] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs) using their voice. For example, a user might say, "I'm looking for a new smartphone." This input is captured by a microphone built into the user's head-mounted display. The voice data is then converted into text data within the device.

[1221] Terminal operation

[1222] The terminal uses speech recognition software to convert the user's voice input into text data. The speech recognition software used is the Python `speech_recognition` library. The converted text data is sent to the server.

[1223] Server Operations

[1224] The server analyzes the received text data using a natural language processing engine (NLP engine). The NLP engine used is the nlptown / bert-base-multilingual-uncased-sentiment model. Based on the analyzed user intent and question content, it queries a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. This generated response text is then sent back to the user's terminal by the server and converted into an action script for the displayed character.

[1225] NPC Controls

[1226] The NPC displayed on the head-mounted display uses text-to-speech software to play back response text received from the server. The pyttsx3 library is used for this purpose. The NPC provides the generated response to the user as audio, and, if necessary, accompanies it with body movements.

[1227] Specific example

[1228] When a user says "I'm looking for a new smartphone" in the virtual store, the system responds as follows:

[1229] Voice input: "I'm looking for a new smartphone."

[1230] Analysis of user intent: "Prefers to view on a smartphone"

[1231] Response generation: "We have a new smartphone here. Would you like to take a look at the latest model?"

[1232] Example of a prompt

[1233] After analyzing the user's voice input as "I'm looking for a new smartphone," the following prompt is entered into the generative AI model:

[1234] "Generate appropriate responses for users looking for a new smartphone."

[1235] This provides a convenient system that allows users to intuitively obtain product information and effectively proceed with the purchase process.

[1236] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1237] Step 1:

[1238] Users wear a head-mounted display in a virtual store and speak to displayed characters (NPCs). During this process, the user's voice input is captured by a microphone built into the head-mounted display.

[1239] Input: User voice input (e.g., "I'm looking for a new smartphone.")

[1240] Output: Captured audio data

[1241] Step 2:

[1242] The device converts the captured audio data into text data using speech recognition software. This speech recognition uses the Python `speech_recognition` library.

[1243] Input: Captured audio data

[1244] Data processing: Text conversion using speech recognition.

[1245] Output: Converted text data (e.g., "I'm looking for a new smartphone.")

[1246] Step 3:

[1247] The terminal sends the converted text data to the server.

[1248] Input: Converted text data

[1249] Output: Text data sent to the server

[1250] Step 4:

[1251] The server analyzes the received text data using a natural language processing engine. This engine uses the nlptown / bert-base-multilingual-uncased-sentiment model.

[1252] Input: Received text data

[1253] Data processing: Analyzing user intent using natural language processing

[1254] Output: Analyzed user intent (e.g., "I want to view this on my smartphone")

[1255] Step 5:

[1256] The server sends a prompt to a generative AI model (e.g., EleutherAI / gpt-neo-2.7B) to generate an appropriate response. The prompt is "Generate an appropriate response for a user looking for a new smartphone."

[1257] Input: Analyzed user intent, prompt message

[1258] Data processing: Prompt-based response generation

[1259] Output: Generated response text (Example: "We have a new smartphone here. Would you like to take a look at the latest model?")

[1260] Step 6:

[1261] The server sends the generated response text to the user's terminal and converts it into a character script for display on the terminal.

[1262] Input: Generated response text

[1263] Output: Converted display character script

[1264] Step 7:

[1265] The terminal uses the converted script to have the displayed character (NPC) play a response using speech synthesis software (e.g., pyttsx3). During this process, speech synthesis generates audio data, which is then provided to the user through the NPC's speaker. The NPC also performs gestures and other actions as needed.

[1266] Input: Converted display character script

[1267] Data processing: Generation of voice data using speech synthesis, and operation instructions based on scripts.

[1268] Output: Voice responses and gestures to the user

[1269] Through these steps, the user experience within the virtual store is significantly improved, allowing users to intuitively obtain the necessary information and proceed with the purchase process smoothly.

[1270] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1271] This invention relates to a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining a generative artificial intelligence system and an emotion engine for NPCs (non-player characters).

[1272] Overall system configuration

[1273] The system processes and analyzes user input within the metaverse, generates appropriate responses using generative AI, and delivers them to the user through NPCs. Furthermore, by incorporating an emotion engine, it recognizes the user's emotional state and generates responses accordingly. The following describes each part of the system, subject by subject.

[1274] User-side operations

[1275] Users speak to NPCs in the metaverse using voice or text. For example, a user might say to an NPC, "I'm looking for new shoes." This input is captured by the user's device (e.g., computer, smartphone, VR device, etc.), and if it's voice input, the audio is converted into text data.

[1276] Operation on the device side

[1277] The user's device converts the captured audio data into text data using speech recognition technology. The converted text data is then sent to the server. A general speech recognition API (e.g., a speech recognition service) is used as the speech recognition method.

[1278] Server-side operations

[1279] The server uses a natural language processing (NLP) engine to analyze text data received from the terminal. The NLP engine understands the user's intent and the content of the question, and generates analysis results. Based on these analysis results, it sends a request to a generative AI (e.g., a natural language generation model) to generate an appropriate response.

[1280] Furthermore, the server uses an emotion engine to analyze the user's voice or text data and recognize the user's emotions. Based on the analysis results of the emotion engine, the generative AI adjusts its response. This process generates a response that is more appropriate to the user's state of mind.

[1281] The generative AI generates appropriate text responses corresponding to user input and the results of the emotion engine, and returns them to the server. The server converts this response text into an action script for an NPC and sends it to the NPC.

[1282] NPC-side operations

[1283] NPCs respond to the user according to the action scripts they receive from the server. These responses are output as audio and provided to the user. NPCs can also perform specific actions or gestures as needed.

[1284] Specific example

[1285] As a concrete example, let's consider the interaction between a user and an NPC in a fashion shop within the metaverse.

[1286] 1. User: Enter a fashion shop in the metaverse and talk to an NPC saying, "I'm looking for new shoes."

[1287] 2. Terminal: Captures the user's voice and converts it into text data, "I'm looking for new shoes," using speech recognition technology. This text is then sent to the server.

[1288] 3. Server: The server analyzes the received text data using an NLP engine and converts the user's intent into information such as "I am looking for shoes." Furthermore, it uses an emotion engine to analyze the user's voice data and recognize emotional states such as "excited" or "distressed."

[1289] 4. Server: Based on the analysis results and the emotion engine's results, it gives instructions to the generative AI to generate responses such as, "We have new shoes here. Would you like to take a look?" or "I'm here to help if you need anything."

[1290] 5. Server: Converts the generated response into an NPC action script and sends it to the NPC.

[1291] 6. NPC: The NPC will say to the user, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf. It can also change its facial expressions and tone of voice in response to the user's emotions.

[1292] 7. User: Follow the NPC's instructions and go to check out the new shoes.

[1293] In this way, the system of the present invention improves the user experience within the metaverse. By combining it with an emotion engine, NPCs can not only respond flexibly to the user's specific questions and intentions, but also provide responses that take into account the user's emotional state.

[1294] The following describes the processing flow.

[1295] Step 1:

[1296] The user speaks to an NPC in the metaverse. For example, the user might say, "I'm looking for new shoes."

[1297] Step 2:

[1298] The device captures the user's voice or text input. In the case of voice input, it collects the voice data and prepares it for processing.

[1299] Step 3:

[1300] The device uses speech recognition to convert speech data into text data. It calls a speech recognition API (e.g., a speech recognition service) to perform the conversion.

[1301] Step 4:

[1302] The terminal sends the converted text data to the server. The process involves sending the text data to the server via network communication.

[1303] Step 5:

[1304] The server receives the text data from the terminal. The received data is then passed on to the next parsing step.

[1305] Step 6:

[1306] The server invokes the NLP engine to analyze the received text data. The NLP engine analyzes the user's intent and the content of the question, and generates the results.

[1307] Step 7:

[1308] The server invokes an emotion engine to analyze the received audio or text data and recognize the user's emotions. For example, it extracts emotions from voice tone or text.

[1309] Step 8:

[1310] The server sends a request to the generative AI to generate a response based on the analysis results of the NLP engine and the emotion engine. For example, it provides context such as "the user is looking for shoes" or "the user is in trouble."

[1311] Step 9:

[1312] A generative AI receives a request from a server and generates an appropriate response. For example, it might generate a response like, "We have new shoes here. Would you like to take a look?"

[1313] Step 10:

[1314] The server receives the response text from the generative AI. The received text is converted into an NPC action script.

[1315] Step 11:

[1316] The server creates an action script for the NPC based on the response text and sends it to the NPC. The action script may also include the NPC's gestures and animations.

[1317] Step 12:

[1318] The NPC responds to the user according to the action script received from the server. For example, it might say, "We have new shoes here. Would you like to take a look?" and then point to a specific shoe shelf.

[1319] Step 13:

[1320] The user acts according to the NPC's instructions. The user approaches the shoe rack indicated by the NPC's response and checks the product.

[1321] Through the processing flow described above, the system of the present invention, including the emotion engine, enables natural interaction and guidance with the user within the metaverse, thereby improving the user experience.

[1322] (Example 2)

[1323] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1324] Traditional metaverse environments have a problem with dialogue systems with NPCs (non-player characters) that degrade the quality of the user experience because they proceed without considering the user's emotional state. Furthermore, it has been difficult to build a system that analyzes user input in real time and generates appropriate responses.

[1325] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1326] In this invention, the server includes means for receiving user input, means for analyzing the input data, means for using a generative artificial intelligence system to generate a response corresponding to the analysis results and the user's emotional state, means for transmitting the generated response to an NPC, and means for the NPC to respond to the user. This makes it possible to consider the user's emotional state and generate and provide an appropriate response in real time.

[1327] "Means of receiving user input" refers to devices or functions that capture and collect user input, whether in voice or text, within a metaverse environment.

[1328] "Means for analyzing input data" refers to software or algorithms that understand and classify the content and intent of user input data received from that user.

[1329] "Means of using generative artificial intelligence" refers to devices or functions that use artificial intelligence technology to generate appropriate responses in real time based on analyzed input data.

[1330] "Means for transmitting generated responses to NPCs" refers to devices or functions for transmitting responses generated by generative artificial intelligence to NPCs in an appropriate format.

[1331] "Means by which NPCs respond to users" refers to devices or functions that enable NPCs to interactively respond to users through voice or actions based on the responses they have generated.

[1332] "A speech recognition method that converts user voice input into text" refers to software or algorithms that convert the voice spoken by a user into digital text data.

[1333] A "natural language processing engine" refers to the technology and programs used to analyze text data and understand its content and meaning.

[1334] An "emotion engine" is software or algorithms that analyze and determine a user's emotional state from their voice and text data.

[1335] This invention is a system that enhances the user experience by enabling real-time interaction with users in a metaverse environment by combining generative artificial intelligence and an emotion engine for NPCs (non-player characters). This system uses the following hardware and software to analyze user input in real time and generate natural responses according to the user's emotional state.

[1336] First, the user approaches an NPC in the metaverse and speaks to them using voice or text. For example, they might say, "I'm looking for new shoes." This input is captured by the user's device (computer, smartphone, VR device, etc.). In the case of voice input, the microphone in the device collects the voice data.

[1337] The device converts the captured audio data into text data using speech recognition (for example, the Google Speech-to-Text API). This process generates text data such as "I'm looking for new shoes." This text data is then sent to the server.

[1338] The server uses a natural language processing (NLP) engine (for example, the Google Cloud Natural Language API) to analyze the received text data. This engine understands the user's intent and question from the text data and generates analysis results such as "looking for shoes."

[1339] Furthermore, the server uses an emotion engine (e.g., Affectiva SDK) to analyze the user's voice or text data and recognize the user's emotional state. For example, it can determine if the user is "excited." Based on this analysis, it sends a prompt to a generative artificial intelligence (e.g., GPT-3) to generate an appropriate response. An example of a prompt might be, "Generate an appropriate response if the user is excited and looking for new shoes."

[1340] The server generates an action script for the NPC based on the response returned by the generative artificial intelligence. This script includes the content of the response to be output as voice and actions such as pointing to a specific product shelf. The server then sends this action script to the NPC.

[1341] The NPC responds to the user according to the action script it receives. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf. It can also adjust its facial expressions and tone of voice according to the user's emotions.

[1342] As a concrete example, let's look at a conversation between a user and an NPC in a fashion shop within the metaverse. When the user speaks aloud, "I'm looking for new shoes," the user's device converts the speech to text and sends it to the server. The server analyzes the received text data and understands that the user is "looking for shoes." Furthermore, if the emotion engine determines that the user is "excited," it instructs GPT-3 to "respond to an excited user" based on this analysis result. The generated response is sent to the NPC, who responds with voice and actions saying, "We have new shoes here. Would you like to take a look?"

[1343] This system can significantly improve the user experience within the metaverse. By using an emotion engine in conjunction with it, NPCs will respond in a way that takes the user's emotional state into account, resulting in more natural and satisfying conversations.

[1344] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1345] Step 1:

[1346] The user speaks to an NPC in the metaverse and says, "I'm looking for new shoes."

[1347] Input: User voice or text input.

[1348] Specific action: The user approaches an NPC through a VR device in the metaverse and speaks to them using voice, saying, "I'm looking for new shoes."

[1349] Step 2:

[1350] The audio data captured by the device is converted into text data using a speech recognition means (e.g., a speech recognition API).

[1351] Input: Captured audio data.

[1352] Output: Converted text data.

[1353] Specific operation: The user's device uses its microphone to recognize speech and converts it into text data, such as "I'm looking for new shoes," via a speech recognition API. This text data is then sent to the server over the internet.

[1354] Step 3:

[1355] The server uses a natural language processing engine (e.g., a natural language processing API) to analyze the text data it receives.

[1356] Input: Text data submitted by the user.

[1357] Output: Identification of intent and content through analysis (e.g., "Looking for shoes").

[1358] Specific operation: The server inputs text data into a natural language processing engine, tags it with the intention "looking for shoes," and analyzes it.

[1359] Step 4:

[1360] The server uses an emotion engine (e.g., an emotion analysis API) to analyze the user's text data and recognize their emotional state.

[1361] Input: User's text data.

[1362] Output: Sentiment analysis result (e.g., "excited").

[1363] Specific operation: The server inputs text data into the emotion engine and determines that the emotion is "excited".

[1364] Step 5:

[1365] The server sends prompts to a generative AI model (e.g., a generative AI model) based on the analysis results and emotional state, and generates an appropriate response.

[1366] Input: Analysis results and emotional state.

[1367] Output: The generated response (for example, "We have new shoes here. Would you like to take a look?").

[1368] Specific operation: The server sends a prompt to the generative AI model saying, "Respond to a user looking for new shoes in an excited manner," and receives the generated response.

[1369] Step 6:

[1370] The server converts the generated response into an action script for the NPC and sends it to the NPC.

[1371] Input: Generated response.

[1372] Output: NPC action script.

[1373] Specific operation: The server converts the response into text-to-speech, generates a script that includes the actions and gestures the NPC should perform, and sends it to the NPC.

[1374] Step 7:

[1375] The NPC responds to the user based on the action script received from the server.

[1376] Input: NPC action script.

[1377] Output: Responses to the user (voice and gestures).

[1378] Specific action: The NPC will say in voice, "We have new shoes here. Would you like to take a look?" while pointing to a product shelf.

[1379] (Application Example 2)

[1380] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1381] In traditional metaverse environments, non-player characters (NPCs) have limited interaction with the user and can only respond based on pre-set scenarios, thus limiting the user experience. Furthermore, they lack the ability to recognize and respond to the user's emotional state, making deep dialogue difficult. As a result, conversations with NPCs often felt unnatural, leading to decreased user engagement.

[1382] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for receiving user input, means for analyzing input data, means for using a generative artificial intelligence system to generate a response based on the analysis results, means for transmitting the generated response to an autonomous agent, means for the autonomous agent to respond to the user, means for recognizing the user's emotional state from the input data using an emotion engine, and means for generating a response corresponding to the user's emotional state. This makes it possible to provide an appropriate and emotionally sensitive response to user input.

[1383] "Means for receiving user input" refers to an interface for receiving voice or text-based input from a user in the metaverse environment.

[1384] "Means for analyzing input data" refers to technologies that analyze data received from users to understand their intentions and content.

[1385] "Means using generative artificial intelligence to generate responses" refers to artificial intelligence technology for automatically generating appropriate responses based on analyzed data.

[1386] "Means for sending to an autonomous agent" refers to a mechanism for sending the generated response to an agent that acts autonomously.

[1387] "A means by which an autonomous agent responds to a user" refers to a method by which an autonomous agent responds to a user based on the content of the response it receives.

[1388] "A means of recognizing a user's emotional state from input data using an emotion engine" refers to an engine that analyzes a user's voice or text to recognize their emotions.

[1389] "Means for generating responses that correspond to the user's emotional state" refers to technologies for generating the optimal response based on recognized emotions.

[1390] This invention is a system for making user interaction in a metaverse environment more realistic and profound, and includes the following configuration and processing steps.

[1391] System Overview

[1392] User actions

[1393] Users wear smart glasses within the metaverse and communicate with NPCs via voice and text. For example, a user might say, "I'm looking for new shoes." This input is captured by the user's device (smart glasses).

[1394] Operation of the device (smart glasses)

[1395] The smart glasses convert captured audio data into text data using a speech recognition API. Google Cloud Speech-to-Text is used as this speech recognition API. This converted text data is then sent to the server.

[1396] Server Operations

[1397] The server analyzes text data received from the terminal using a natural language processing engine (Google Cloud Natural Language). This analysis clarifies the user's intent. It also analyzes emotions from the user's voice and text using an emotion engine. This emotion engine utilizes an emotion analysis API such as IBM Watson. Based on these analysis results, an appropriate response is generated using a generative AI model (OpenAI GPT-4).

[1398] Controlling Autonomous Agents (NPCs)

[1399] The generated response is sent from the server to the autonomous agent. The autonomous agent provides an appropriate response to the user through voice and actions. For example, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a specific product shelf.

[1400] Specific example

[1401] 1. User: Puts on smart glasses and enters a virtual store in the metaverse. Says, "I'm looking for new shoes."

[1402] 2. Device (smart glasses): Captures the user's voice, converts it into text data such as "I'm looking for new shoes" using Google Cloud Speech-to-Text, and sends it to the server.

[1403] 3. Server: Receives text data and parses it using Google Cloud Natural Language. Understands the user's intent and analyzes the user's emotional state using an emotion engine such as IBM Watson. For example, it might recognize that the user is "excited." Based on this, it uses OpenAI GPT-4 to generate a response such as, "We have new shoes here. Would you like to take a look?"

[1404] 4. Autonomous Agent (NPC): Based on the generated response, it will approach the user and say, "We have new shoes here. Would you like to take a look?" and point to the product shelf.

[1405] Example of a prompt

[1406] Prompt text to input to the generative AI model:

[1407] The user says, "I'm looking for new shoes." The user is excited. Generate an appropriate response.

[1408] This invention provides appropriate and emotionally sensitive responses to user input, improving the user experience within the metaverse.

[1409] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1410] Processing flow of the system program that implements the application example

[1411] Step 1:

[1412] The user wears smart glasses and uses voice input within the metaverse. For example, they might say, "I'm looking for new shoes." This voice data is captured by the microphone built into the smart glasses.

[1413] Input: User's voice data

[1414] Output: Audio data

[1415] Step 2:

[1416] The device (smart glasses) converts the captured audio data into text data using a speech recognition API (Google Cloud Speech-to-Text).

[1417] Input: Audio data

[1418] Data processing: Convert audio data to text data using the Google Cloud Speech-to-Text API.

[1419] Output: Text data

[1420] Step 3:

[1421] The terminal sends the converted text data to the server.

[1422] Input: Text data

[1423] Output: Text data sent to the server

[1424] Step 4:

[1425] The server uses a natural language processing engine (Google Cloud Natural Language) to analyze text data and understand the user's intent.

[1426] Input: Text data

[1427] Data processing: Analyze user intent using Google Cloud Natural Language.

[1428] Output: User intent

[1429] Step 5:

[1430] The server uses an emotion engine (such as IBM Watson) to analyze the user's emotional state. This analysis identifies emotions from text data.

[1431] Input: Text data

[1432] Data processing: Analyze user emotions using an emotion engine.

[1433] Output: User's emotional state (e.g., excited)

[1434] Step 6:

[1435] The server uses a generative AI model (OpenAI GPT-4) to generate an appropriate response based on the analyzed user intent and emotional state. For example, it can take the prompt "The user said, 'I'm looking for new shoes.' The user is in an excited state. Generate an appropriate response." and generate a response.

[1436] Input: User intent, emotional state, prompt text

[1437] Data processing: Generate responses using OpenAI GPT-4.

[1438] Output: Generated response (Example: "We have new shoes here. Would you like to take a look?")

[1439] Step 7:

[1440] The server converts the generated response into an NPC action script and sends that script to the autonomous agent (NPC).

[1441] Input: Generated response

[1442] Data processing: Convert response text into action scripts

[1443] Output: Action script for NPCs

[1444] Step 8:

[1445] The autonomous agent (NPC) responds to the user with voice and actions according to an action script. Specifically, it might say, "We have new shoes here. Would you like to take a look?" while pointing to a shelf.

[1446] Input: Action script

[1447] Specific actions: The customer responds with a voice, pointing to a shelf and saying, "We have new shoes here. Would you like to take a look?"

[1448] Output: Response to the user

[1449] In this way, appropriate and emotionally sensitive responses based on user input are provided at every step.

[1450] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1451] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1452] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1453] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1454] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1455] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1456] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1457] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1458] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1459] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1460] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1461] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1462] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1463] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1464] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1465] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1466] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1467] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1468] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1469] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1470] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[1471] The following is further disclosed regarding the embodiments described above.

[1472] (Claim 1)

[1473] A means of receiving user input,

[1474] A means of analyzing input data,

[1475] A method using a generative artificial intelligence that generates a response based on the analysis results,

[1476] A means of sending the generated response to an NPC (non-player character),

[1477] A means by which NPCs respond to users,

[1478] A system that includes this.

[1479] (Claim 2)

[1480] The system according to claim 1, further comprising speech recognition means for converting user voice input into text.

[1481] (Claim 3)

[1482] The system according to claim 1, wherein a natural language processing engine is used as the analysis means.

[1483] "Example 1"

[1484] (Claim 1)

[1485] Means for obtaining user voice or text input,

[1486] A speech recognition means that converts voice input into text data,

[1487] A method using a natural language processing engine to analyze acquired text data,

[1488] A means of using a generative artificial intelligence that generates an appropriate response based on the analysis results,

[1489] A means of converting the generated response into an NPC action script and sending it to the NPC,

[1490] A means by which NPCs respond to users,

[1491] A system that includes this.

[1492] (Claim 2)

[1493] The system according to claim 1, further comprising means for converting acquired audio data into text data and transmitting it to a server.

[1494] (Claim 3)

[1495] The system according to claim 1, further comprising means for an NPC to perform a specific action or gesture based on the generated response.

[1496] "Application Example 1"

[1497] (Claim 1)

[1498] A means of receiving user input,

[1499] A means of analyzing input data,

[1500] A method using a generative artificial intelligence that generates a response based on the analysis results,

[1501] A means for sending the generated response to the display character,

[1502] A means by which the displayed character responds to the user,

[1503] A means for speech recognition and speech synthesis for voice input and voice output,

[1504] A system that includes this.

[1505] (Claim 2)

[1506] The system according to claim 1, further comprising speech recognition means for converting user voice input into text.

[1507] (Claim 3)

[1508] The system according to claim 1, wherein a natural language processing engine is used as the analysis means.

[1509] "Example 2 of combining an emotion engine"

[1510] (Claim 1)

[1511] A means of receiving user input,

[1512] A means of analyzing input data,

[1513] A method using generative artificial intelligence that generates responses according to the analysis results and the user's emotional state,

[1514] A means of sending the generated response to the NPC,

[1515] A means by which NPCs respond to users,

[1516] A system that includes this.

[1517] (Claim 2)

[1518] The system according to claim 1, further comprising speech recognition means for converting user voice input into text.

[1519] (Claim 3)

[1520] The system according to claim 1, wherein a natural language processing engine and an emotion engine are used as analytical means.

[1521] "Application example 2 when combining with an emotional engine"

[1522] (Claim 1)

[1523] A means of receiving user input,

[1524] A means of analyzing input data,

[1525] A method using a generative artificial intelligence that generates a response based on the analysis results,

[1526] A means for sending the generated response to an autonomous agent,

[1527] The means by which an autonomous agent responds to the user,

[1528] A means of recognizing a user's emotional state from input data using an emotion engine,

[1529] A means for generating responses that correspond to the user's emotional state,

[1530] A system that includes this.

[1531] (Claim 2)

[1532] The system according to claim 1, further comprising speech recognition means for converting user voice input into text.

[1533] (Claim 3)

[1534] The system according to claim 1, wherein a natural language processing engine is used as the analysis means. [Explanation of Symbols]

[1535] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving user input, A means of analyzing input data, A method using a generative artificial intelligence that generates a response based on the analysis results, A means of sending the generated response to the NPC, A means by which NPCs respond to users, A system that includes this.

2. The system according to claim 1, further comprising speech recognition means for converting user voice input into text.

3. The system according to claim 1, wherein a natural language processing engine is used as the analysis means.

Images